# Data cleaning and eda
import warnings
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
from scipy import stats

# Data processing and data augmentation
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
import matplotlib.gridspec as gridspec
from imblearn.over_sampling import SMOTE
from sklearn.feature_selection import mutual_info_classif

# Model general
from sklearn.metrics import accuracy_score, f1_score, confusion_matrix, ConfusionMatrixDisplay, precision_score, recall_score, roc_auc_score
from sklearn.model_selection import cross_val_score, GridSearchCV
from sklearn.model_selection import GridSearchCV, RandomizedSearchCV

# SVM
from sklearn.svm import SVC

# Random forest
from sklearn.ensemble import RandomForestClassifier

# CNN
# !pip install tensorflow
import tensorflow as tf

# Shap
!pip install shap
import shap

warnings.filterwarnings("ignore")

Collecting shap
  Downloading shap-0.43.0-cp310-cp310-manylinux_2_12_x86_64.manylinux2010_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl (532 kB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 532.9/532.9 kB 4.7 MB/s eta 0:00:00
Requirement already satisfied: numpy in /usr/local/lib/python3.10/dist-packages (from shap) (1.23.5)
Requirement already satisfied: scipy in /usr/local/lib/python3.10/dist-packages (from shap) (1.11.3)
Requirement already satisfied: scikit-learn in /usr/local/lib/python3.10/dist-packages (from shap) (1.2.2)
Requirement already satisfied: pandas in /usr/local/lib/python3.10/dist-packages (from shap) (1.5.3)
Requirement already satisfied: tqdm>=4.27.0 in /usr/local/lib/python3.10/dist-packages (from shap) (4.66.1)
Requirement already satisfied: packaging>20.9 in /usr/local/lib/python3.10/dist-packages (from shap) (23.2)
Collecting slicer==0.0.7 (from shap)
  Downloading slicer-0.0.7-py3-none-any.whl (14 kB)
Requirement already satisfied: numba in /usr/local/lib/python3.10/dist-packages (from shap) (0.56.4)
Requirement already satisfied: cloudpickle in /usr/local/lib/python3.10/dist-packages (from shap) (2.2.1)
Requirement already satisfied: llvmlite<0.40,>=0.39.0dev0 in /usr/local/lib/python3.10/dist-packages (from numba->shap) (0.39.1)
Requirement already satisfied: setuptools in /usr/local/lib/python3.10/dist-packages (from numba->shap) (67.7.2)
Requirement already satisfied: python-dateutil>=2.8.1 in /usr/local/lib/python3.10/dist-packages (from pandas->shap) (2.8.2)
Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.10/dist-packages (from pandas->shap) (2023.3.post1)
Requirement already satisfied: joblib>=1.1.1 in /usr/local/lib/python3.10/dist-packages (from scikit-learn->shap) (1.3.2)
Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.10/dist-packages (from scikit-learn->shap) (3.2.0)
Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.10/dist-packages (from python-dateutil>=2.8.1->pandas->shap) (1.16.0)
Installing collected packages: slicer, shap
Successfully installed shap-0.43.0 slicer-0.0.7

from google.colab import drive
drive.mount('/content/drive')
dataset = pd.read_csv('/content/drive/My Drive/BT4012 Fraud Analytics/Code/transaction_dataset.csv')
dataset.head()

Mounted at /content/drive

print("The shape of the dataset is: " + str(dataset.shape))

The shape of the dataset is: (9841, 51)

dataset.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 9841 entries, 0 to 9840
Data columns (total 51 columns):
 #   Column                                                Non-Null Count  Dtype  
---  ------                                                --------------  -----  
 0   Unnamed: 0                                            9841 non-null   int64  
 1   Index                                                 9841 non-null   int64  
 2   Address                                               9841 non-null   object 
 3   FLAG                                                  9841 non-null   int64  
 4   Avg min between sent tnx                              9841 non-null   float64
 5   Avg min between received tnx                          9841 non-null   float64
 6   Time Diff between first and last (Mins)               9841 non-null   float64
 7   Sent tnx                                              9841 non-null   int64  
 8   Received Tnx                                          9841 non-null   int64  
 9   Number of Created Contracts                           9841 non-null   int64  
 10  Unique Received From Addresses                        9841 non-null   int64  
 11  Unique Sent To Addresses                              9841 non-null   int64  
 12  min value received                                    9841 non-null   float64
 13  max value received                                    9841 non-null   float64
 14  avg val received                                      9841 non-null   float64
 15  min val sent                                          9841 non-null   float64
 16  max val sent                                          9841 non-null   float64
 17  avg val sent                                          9841 non-null   float64
 18  min value sent to contract                            9841 non-null   float64
 19  max val sent to contract                              9841 non-null   float64
 20  avg value sent to contract                            9841 non-null   float64
 21  total transactions (including tnx to create contract  9841 non-null   int64  
 22  total Ether sent                                      9841 non-null   float64
 23  total ether received                                  9841 non-null   float64
 24  total ether sent contracts                            9841 non-null   float64
 25  total ether balance                                   9841 non-null   float64
 26   Total ERC20 tnxs                                     9012 non-null   float64
 27   ERC20 total Ether received                           9012 non-null   float64
 28   ERC20 total ether sent                               9012 non-null   float64
 29   ERC20 total Ether sent contract                      9012 non-null   float64
 30   ERC20 uniq sent addr                                 9012 non-null   float64
 31   ERC20 uniq rec addr                                  9012 non-null   float64
 32   ERC20 uniq sent addr.1                               9012 non-null   float64
 33   ERC20 uniq rec contract addr                         9012 non-null   float64
 34   ERC20 avg time between sent tnx                      9012 non-null   float64
 35   ERC20 avg time between rec tnx                       9012 non-null   float64
 36   ERC20 avg time between rec 2 tnx                     9012 non-null   float64
 37   ERC20 avg time between contract tnx                  9012 non-null   float64
 38   ERC20 min val rec                                    9012 non-null   float64
 39   ERC20 max val rec                                    9012 non-null   float64
 40   ERC20 avg val rec                                    9012 non-null   float64
 41   ERC20 min val sent                                   9012 non-null   float64
 42   ERC20 max val sent                                   9012 non-null   float64
 43   ERC20 avg val sent                                   9012 non-null   float64
 44   ERC20 min val sent contract                          9012 non-null   float64
 45   ERC20 max val sent contract                          9012 non-null   float64
 46   ERC20 avg val sent contract                          9012 non-null   float64
 47   ERC20 uniq sent token name                           9012 non-null   float64
 48   ERC20 uniq rec token name                            9012 non-null   float64
 49   ERC20 most sent token type                           9000 non-null   object 
 50   ERC20_most_rec_token_type                            8990 non-null   object 
dtypes: float64(39), int64(9), object(3)
memory usage: 3.8+ MB

duplicate_rows = dataset[dataset.duplicated()]
num_duplicated_rows = duplicate_rows.shape[0]
print(f"The number of duplicate rows in this dataset is: {num_duplicated_rows}")

The number of duplicate rows in this dataset is: 0

# Gives total count of fraudulent transactions
fraudulent_count = dataset['FLAG'].value_counts()[1]

# Gives total count of non-fraudulent transactions
non_fraudulent_count = dataset['FLAG'].value_counts()[0]

# Required ratio
ratio = fraudulent_count/non_fraudulent_count

# Answer
print("The ratio of fraud vs non-fraud transactions is " + str(ratio) +'.')

The ratio of fraud vs non-fraud transactions is 0.2843904985643435.

# Drop index-related columns that are not useful
dataset.drop(columns=['Unnamed: 0', "Index", 'Address'], axis=1, inplace=True)

# Look at dataset again
dataset.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 9841 entries, 0 to 9840
Data columns (total 48 columns):
 #   Column                                                Non-Null Count  Dtype  
---  ------                                                --------------  -----  
 0   FLAG                                                  9841 non-null   int64  
 1   Avg min between sent tnx                              9841 non-null   float64
 2   Avg min between received tnx                          9841 non-null   float64
 3   Time Diff between first and last (Mins)               9841 non-null   float64
 4   Sent tnx                                              9841 non-null   int64  
 5   Received Tnx                                          9841 non-null   int64  
 6   Number of Created Contracts                           9841 non-null   int64  
 7   Unique Received From Addresses                        9841 non-null   int64  
 8   Unique Sent To Addresses                              9841 non-null   int64  
 9   min value received                                    9841 non-null   float64
 10  max value received                                    9841 non-null   float64
 11  avg val received                                      9841 non-null   float64
 12  min val sent                                          9841 non-null   float64
 13  max val sent                                          9841 non-null   float64
 14  avg val sent                                          9841 non-null   float64
 15  min value sent to contract                            9841 non-null   float64
 16  max val sent to contract                              9841 non-null   float64
 17  avg value sent to contract                            9841 non-null   float64
 18  total transactions (including tnx to create contract  9841 non-null   int64  
 19  total Ether sent                                      9841 non-null   float64
 20  total ether received                                  9841 non-null   float64
 21  total ether sent contracts                            9841 non-null   float64
 22  total ether balance                                   9841 non-null   float64
 23   Total ERC20 tnxs                                     9012 non-null   float64
 24   ERC20 total Ether received                           9012 non-null   float64
 25   ERC20 total ether sent                               9012 non-null   float64
 26   ERC20 total Ether sent contract                      9012 non-null   float64
 27   ERC20 uniq sent addr                                 9012 non-null   float64
 28   ERC20 uniq rec addr                                  9012 non-null   float64
 29   ERC20 uniq sent addr.1                               9012 non-null   float64
 30   ERC20 uniq rec contract addr                         9012 non-null   float64
 31   ERC20 avg time between sent tnx                      9012 non-null   float64
 32   ERC20 avg time between rec tnx                       9012 non-null   float64
 33   ERC20 avg time between rec 2 tnx                     9012 non-null   float64
 34   ERC20 avg time between contract tnx                  9012 non-null   float64
 35   ERC20 min val rec                                    9012 non-null   float64
 36   ERC20 max val rec                                    9012 non-null   float64
 37   ERC20 avg val rec                                    9012 non-null   float64
 38   ERC20 min val sent                                   9012 non-null   float64
 39   ERC20 max val sent                                   9012 non-null   float64
 40   ERC20 avg val sent                                   9012 non-null   float64
 41   ERC20 min val sent contract                          9012 non-null   float64
 42   ERC20 max val sent contract                          9012 non-null   float64
 43   ERC20 avg val sent contract                          9012 non-null   float64
 44   ERC20 uniq sent token name                           9012 non-null   float64
 45   ERC20 uniq rec token name                            9012 non-null   float64
 46   ERC20 most sent token type                           9000 non-null   object 
 47   ERC20_most_rec_token_type                            8990 non-null   object 
dtypes: float64(39), int64(7), object(2)
memory usage: 3.6+ MB

# Check if first set of columns are duplicates
first_set_columns = dataset[' ERC20 avg time between rec tnx'].equals(dataset[' ERC20 avg time between rec 2 tnx'])

if first_set_columns:
    print(f"The columns ' ERC20 avg time between rec tnx' and ' ERC20 avg time between rec 2 tnx' are exactly the same.")
else:
    print(f"The columns ' ERC20 avg time between rec tnx' and ' ERC20 avg time between rec 2 tnx' are not the same.")

# Check if second set of columns are duplicates
second_set_columns = dataset[' ERC20 uniq sent addr'].equals(dataset[' ERC20 uniq sent addr.1'])

if second_set_columns:
    print(f"The columns ' ERC20 uniq sent addr' and ' ERC20 uniq sent addr.1' are exactly the same.")
else:
    print(f"The columns ' ERC20 uniq sent addr' and ' ERC20 uniq sent addr.1' are not the same.")

The columns ' ERC20 avg time between rec tnx' and ' ERC20 avg time between rec 2 tnx' are exactly the same.
The columns ' ERC20 uniq sent addr' and ' ERC20 uniq sent addr.1' are not the same.

# If the column is indeed ''
filtered_df = dataset[dataset[' ERC20 uniq sent addr.1'] == 0]

# Check if all corresponding values in ' ERC20 total Ether sent contract' are 0
if (filtered_df[' ERC20 total Ether sent contract'] == 0).all():
    print("All values in ' ERC20 total Ether sent contract' are 0 when ' ERC20 uniq sent addr.1' is 0.")
else:
    print("Not all values in ' ERC20 total Ether sent contract' are 0 when ' ERC20 uniq sent addr.1' is 0.")


filtered_df2 = dataset[dataset[' ERC20 uniq sent addr.1'] != 0]

# Check if all corresponding values in ' ERC20 total Ether sent contract' are not 0
if (filtered_df2[' ERC20 total Ether sent contract'] != 0).all():
    print("All values in ' ERC20 total Ether sent contract' are not 0 when ' ERC20 uniq sent addr.1' is not 0.")
else:
    print("Not all values in ' ERC20 total Ether sent contract' are not 0 when ' ERC20 uniq sent addr.1' is not 0.")

All values in ' ERC20 total Ether sent contract' are 0 when ' ERC20 uniq sent addr.1' is 0.
All values in ' ERC20 total Ether sent contract' are not 0 when ' ERC20 uniq sent addr.1' is not 0.

# Rename col 29 to ERC20 uniq sent contract addr
dataset = dataset.rename(columns={' ERC20 uniq sent addr.1':'ERC20 uniq sent contract addr'})

# Drop duplicate columns
dataset.drop(columns=[' ERC20 avg time between rec 2 tnx'], axis=1, inplace=True)

# Rename target variable as Y
dataset = dataset.rename(columns={'FLAG':'Y'})

# Count number of missing values for each column, sort them in descending order
top_missing_columns = dataset.isnull().sum().sort_values(ascending=False)
top_missing_columns

 ERC20_most_rec_token_type                              851
 ERC20 most sent token type                             841
 ERC20 max val rec                                      829
 ERC20 total Ether sent contract                        829
 ERC20 uniq sent addr                                   829
 ERC20 uniq rec addr                                    829
ERC20 uniq sent contract addr                           829
 ERC20 uniq rec contract addr                           829
 ERC20 avg time between sent tnx                        829
 ERC20 avg time between rec tnx                         829
 ERC20 avg time between contract tnx                    829
 ERC20 min val rec                                      829
 ERC20 avg val rec                                      829
 ERC20 total Ether received                             829
 ERC20 min val sent                                     829
 ERC20 max val sent                                     829
 ERC20 avg val sent                                     829
 ERC20 min val sent contract                            829
 ERC20 max val sent contract                            829
 ERC20 avg val sent contract                            829
 ERC20 uniq sent token name                             829
 ERC20 uniq rec token name                              829
 ERC20 total ether sent                                 829
 Total ERC20 tnxs                                       829
Avg min between sent tnx                                  0
avg val received                                          0
Avg min between received tnx                              0
Time Diff between first and last (Mins)                   0
Sent tnx                                                  0
Received Tnx                                              0
Number of Created Contracts                               0
Unique Received From Addresses                            0
Unique Sent To Addresses                                  0
min value received                                        0
max value received                                        0
min val sent                                              0
total ether balance                                       0
max val sent                                              0
avg val sent                                              0
min value sent to contract                                0
max val sent to contract                                  0
avg value sent to contract                                0
total transactions (including tnx to create contract      0
total Ether sent                                          0
total ether received                                      0
total ether sent contracts                                0
Y                                                         0
dtype: int64

# Inspect values of ERC20 most sent token type to decide how to impute values
dataset[' ERC20 most sent token type'].value_counts()

0                                                         4399
None                                                      1856
                                                          1191
EOS                                                        138
OmiseGO                                                    137
                                                          ... 
Arcona Distribution Contract                                 1
HeroCoin                                                     1
Cindicator                                                   1
UnlimitedIP Token                                            1
eosDAC Community Owned EOS Block Producer ERC20 Tokens       1
Name:  ERC20 most sent token type, Length: 305, dtype: int64

# Mapping function to map 'ERC 20 most sent token type'
def map_token_type(token_name):
    if (token_name in ['0', "''"]):
        return 0
    else:
        return 1

# Replace empty values with 0
dataset[' ERC20 most sent token type'].fillna(0, inplace=True)
# Map remainder of the values
dataset[' ERC20 most sent token type'] = dataset[' ERC20 most sent token type'].apply(map_token_type)
# Rename column
dataset = dataset.rename(columns={' ERC20 most sent token type': 'ERC20_most_sent_token_valid_name'})

# Nature of column ' ERC20_most_rec_token_type' is similar to previous case, apply same transformation
# Replace empty values with 0
dataset[' ERC20_most_rec_token_type'].fillna(0, inplace=True)
# Map remainder of the values
dataset[' ERC20_most_rec_token_type'] = dataset[' ERC20_most_rec_token_type'].apply(map_token_type)
# Rename column
dataset = dataset.rename(columns={' ERC20_most_rec_token_type': 'ERC20_most_rec_token_valid_name'})

dataset.isnull().sum().sort_values(ascending=False)

 Total ERC20 tnxs                                       829
 ERC20 max val rec                                      829
 ERC20 total Ether sent contract                        829
 ERC20 uniq sent addr                                   829
 ERC20 uniq rec addr                                    829
ERC20 uniq sent contract addr                           829
 ERC20 uniq rec contract addr                           829
 ERC20 avg time between sent tnx                        829
 ERC20 avg time between rec tnx                         829
 ERC20 avg time between contract tnx                    829
 ERC20 min val rec                                      829
 ERC20 avg val rec                                      829
 ERC20 total Ether received                             829
 ERC20 min val sent                                     829
 ERC20 max val sent                                     829
 ERC20 avg val sent                                     829
 ERC20 min val sent contract                            829
 ERC20 max val sent contract                            829
 ERC20 avg val sent contract                            829
 ERC20 uniq sent token name                             829
 ERC20 uniq rec token name                              829
 ERC20 total ether sent                                 829
ERC20_most_sent_token_valid_name                          0
Y                                                         0
Avg min between sent tnx                                  0
avg val received                                          0
Avg min between received tnx                              0
Time Diff between first and last (Mins)                   0
Sent tnx                                                  0
Received Tnx                                              0
Number of Created Contracts                               0
Unique Received From Addresses                            0
Unique Sent To Addresses                                  0
min value received                                        0
max value received                                        0
min val sent                                              0
total ether balance                                       0
max val sent                                              0
avg val sent                                              0
min value sent to contract                                0
max val sent to contract                                  0
avg value sent to contract                                0
total transactions (including tnx to create contract      0
total Ether sent                                          0
total ether received                                      0
total ether sent contracts                                0
ERC20_most_rec_token_valid_name                           0
dtype: int64

dataset_with_na_rows = dataset[dataset.isnull().any(axis = 1)]
dataset_with_na_rows[[' ERC20 max val rec', ' ERC20 uniq sent addr', ' ERC20 uniq rec addr', 'ERC20 uniq sent contract addr', ' ERC20 uniq rec contract addr',
' ERC20 avg time between sent tnx', ' ERC20 avg time between rec tnx', ' ERC20 avg time between contract tnx', ' ERC20 min val rec',
' ERC20 avg val rec', ' ERC20 total ether sent', ' ERC20 min val sent', ' ERC20 max val sent' , ' ERC20 avg val sent',
' ERC20 min val sent contract', ' ERC20 max val sent contract',' ERC20 avg val sent contract', ' ERC20 uniq sent token name', ' ERC20 uniq rec token name',
' ERC20 total Ether sent contract', ' ERC20 total Ether received',' Total ERC20 tnxs']]

print("Number of rows with missing values for fraud transactions: ",dataset_with_na_rows[dataset_with_na_rows["Y"] == 1].shape[0])

Number of rows with missing values for fraud transactions:  829

# Fraud transactions with no ERC tokens are not identified, as min transaction = 1 instead of 0
min(dataset[dataset["Y"] == 1][' Total ERC20 tnxs'].dropna())

1.0

to_be_imputed = [' ERC20 max val rec', ' ERC20 uniq sent addr', ' ERC20 uniq rec addr', 'ERC20 uniq sent contract addr', ' ERC20 uniq rec contract addr',
' ERC20 avg time between sent tnx', ' ERC20 avg time between rec tnx', ' ERC20 avg time between contract tnx', ' ERC20 min val rec',
' ERC20 avg val rec', ' ERC20 total ether sent', ' ERC20 min val sent', ' ERC20 max val sent' , ' ERC20 avg val sent',
' ERC20 min val sent contract', ' ERC20 max val sent contract',' ERC20 avg val sent contract', ' ERC20 uniq sent token name', ' ERC20 uniq rec token name',
' ERC20 total Ether sent contract', ' ERC20 total Ether received',' Total ERC20 tnxs']
for col in to_be_imputed:
  dataset[col].fillna(0, inplace=True)

# Inspect resultant df to check that there are not more missing values
dataset.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 9841 entries, 0 to 9840
Data columns (total 47 columns):
 #   Column                                                Non-Null Count  Dtype  
---  ------                                                --------------  -----  
 0   Y                                                     9841 non-null   int64  
 1   Avg min between sent tnx                              9841 non-null   float64
 2   Avg min between received tnx                          9841 non-null   float64
 3   Time Diff between first and last (Mins)               9841 non-null   float64
 4   Sent tnx                                              9841 non-null   int64  
 5   Received Tnx                                          9841 non-null   int64  
 6   Number of Created Contracts                           9841 non-null   int64  
 7   Unique Received From Addresses                        9841 non-null   int64  
 8   Unique Sent To Addresses                              9841 non-null   int64  
 9   min value received                                    9841 non-null   float64
 10  max value received                                    9841 non-null   float64
 11  avg val received                                      9841 non-null   float64
 12  min val sent                                          9841 non-null   float64
 13  max val sent                                          9841 non-null   float64
 14  avg val sent                                          9841 non-null   float64
 15  min value sent to contract                            9841 non-null   float64
 16  max val sent to contract                              9841 non-null   float64
 17  avg value sent to contract                            9841 non-null   float64
 18  total transactions (including tnx to create contract  9841 non-null   int64  
 19  total Ether sent                                      9841 non-null   float64
 20  total ether received                                  9841 non-null   float64
 21  total ether sent contracts                            9841 non-null   float64
 22  total ether balance                                   9841 non-null   float64
 23   Total ERC20 tnxs                                     9841 non-null   float64
 24   ERC20 total Ether received                           9841 non-null   float64
 25   ERC20 total ether sent                               9841 non-null   float64
 26   ERC20 total Ether sent contract                      9841 non-null   float64
 27   ERC20 uniq sent addr                                 9841 non-null   float64
 28   ERC20 uniq rec addr                                  9841 non-null   float64
 29  ERC20 uniq sent contract addr                         9841 non-null   float64
 30   ERC20 uniq rec contract addr                         9841 non-null   float64
 31   ERC20 avg time between sent tnx                      9841 non-null   float64
 32   ERC20 avg time between rec tnx                       9841 non-null   float64
 33   ERC20 avg time between contract tnx                  9841 non-null   float64
 34   ERC20 min val rec                                    9841 non-null   float64
 35   ERC20 max val rec                                    9841 non-null   float64
 36   ERC20 avg val rec                                    9841 non-null   float64
 37   ERC20 min val sent                                   9841 non-null   float64
 38   ERC20 max val sent                                   9841 non-null   float64
 39   ERC20 avg val sent                                   9841 non-null   float64
 40   ERC20 min val sent contract                          9841 non-null   float64
 41   ERC20 max val sent contract                          9841 non-null   float64
 42   ERC20 avg val sent contract                          9841 non-null   float64
 43   ERC20 uniq sent token name                           9841 non-null   float64
 44   ERC20 uniq rec token name                            9841 non-null   float64
 45  ERC20_most_sent_token_valid_name                      9841 non-null   int64  
 46  ERC20_most_rec_token_valid_name                       9841 non-null   int64  
dtypes: float64(38), int64(9)
memory usage: 3.5 MB

# Check for less or zero variance
numeric_columns = dataset.select_dtypes(include=['int64', 'float64'])

# Calculate the variance for each numeric column
for column in numeric_columns.columns:
    column_variance = numeric_columns[column].var()
    print(f"Variance in '{column}': {column_variance}")

Variance in 'Y': 0.17241103050701273
Variance in 'Avg min between sent tnx': 461671829.791403
Variance in 'Avg min between received tnx': 532765558.15345657
Variance in 'Time Diff between first and last (Mins)': 104288903881.01013
Variance in 'Sent tnx': 573391.7611902419
Variance in 'Received Tnx': 885173.4141139443
Variance in 'Number of Created Contracts': 20006.852947970285
Variance in 'Unique Received From Addresses': 89174.56869358987
Variance in 'Unique Sent To Addresses': 69601.20869607243
Variance in 'min value received': 106229.80395792934
Variance in 'max value received ': 169229437.82202128
Variance in 'avg val received': 8323237.903808778
Variance in 'min val sent': 19212.643822774065
Variance in 'max val sent': 43946460.26083183
Variance in 'avg val sent': 57159.349305730815
Variance in 'min value sent to contract': 5.0803714012142645e-08
Variance in 'max val sent to contract': 2.6606518910266735e-07
Variance in 'avg value sent to contract': 1.0460961450311993e-07
Variance in 'total transactions (including tnx to create contract': 1828996.6146157237
Variance in 'total Ether sent': 128395165629.00435
Variance in 'total ether received': 132645116719.714
Variance in 'total ether sent contracts': 2.660625045367055e-07
Variance in 'total ether balance': 58770085659.43741
Variance in ' Total ERC20 tnxs': 183510.17473757546
Variance in ' ERC20 total Ether received': 1.017063241588219e+20
Variance in ' ERC20 total ether sent': 1.2759509660733422e+18
Variance in ' ERC20 total Ether sent contract': 34396748.246553466
Variance in ' ERC20 uniq sent addr': 10147.234678398734
Variance in ' ERC20 uniq rec addr': 6134.738618051409
Variance in 'ERC20 uniq sent contract addr': 0.0039534905815474175
Variance in ' ERC20 uniq rec contract addr': 274.2391314791361
Variance in ' ERC20 avg time between sent tnx': 0.0
Variance in ' ERC20 avg time between rec tnx': 0.0
Variance in ' ERC20 avg time between contract tnx': 0.0
Variance in ' ERC20 min val rec': 261048823.60781482
Variance in ' ERC20 max val rec': 1.016835284246567e+20
Variance in ' ERC20 avg val rec': 4.1985986381545176e+16
Variance in ' ERC20 min val sent': 1016498764325.933
Variance in ' ERC20 max val sent': 1.274901251345477e+18
Variance in ' ERC20 avg val sent': 3.2037375380588045e+17
Variance in ' ERC20 min val sent contract': 0.0
Variance in ' ERC20 max val sent contract': 0.0
Variance in ' ERC20 avg val sent contract': 0.0
Variance in ' ERC20 uniq sent token name': 41.688188952309204
Variance in ' ERC20 uniq rec token name': 256.537539871766
Variance in 'ERC20_most_sent_token_valid_name': 0.24722766788688105
Variance in 'ERC20_most_rec_token_valid_name': 0.24721690736366767

# Drop columns with zero or extremely low variance
dataset.drop(columns=[' ERC20 avg time between sent tnx', ' ERC20 avg time between rec tnx', ' ERC20 avg time between contract tnx',
' ERC20 min val sent contract', ' ERC20 max val sent contract', ' ERC20 avg val sent contract', 'min value sent to contract', 'max val sent to contract',
'avg value sent to contract', 'total ether sent contracts', 'ERC20 uniq sent contract addr'], axis=1, inplace=True)

dataset[' ERC20 avg val sent'].value_counts()

0.000000        8373
100.000000       121
0.000001          11
10000.000000       5
1.000000           4
                ... 
0.938440           1
0.452260           1
586.270104         1
112.555556         1
18.571500          1
Name:  ERC20 avg val sent, Length: 1309, dtype: int64

dataset.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 9841 entries, 0 to 9840
Data columns (total 36 columns):
 #   Column                                                Non-Null Count  Dtype  
---  ------                                                --------------  -----  
 0   Y                                                     9841 non-null   int64  
 1   Avg min between sent tnx                              9841 non-null   float64
 2   Avg min between received tnx                          9841 non-null   float64
 3   Time Diff between first and last (Mins)               9841 non-null   float64
 4   Sent tnx                                              9841 non-null   int64  
 5   Received Tnx                                          9841 non-null   int64  
 6   Number of Created Contracts                           9841 non-null   int64  
 7   Unique Received From Addresses                        9841 non-null   int64  
 8   Unique Sent To Addresses                              9841 non-null   int64  
 9   min value received                                    9841 non-null   float64
 10  max value received                                    9841 non-null   float64
 11  avg val received                                      9841 non-null   float64
 12  min val sent                                          9841 non-null   float64
 13  max val sent                                          9841 non-null   float64
 14  avg val sent                                          9841 non-null   float64
 15  total transactions (including tnx to create contract  9841 non-null   int64  
 16  total Ether sent                                      9841 non-null   float64
 17  total ether received                                  9841 non-null   float64
 18  total ether balance                                   9841 non-null   float64
 19   Total ERC20 tnxs                                     9841 non-null   float64
 20   ERC20 total Ether received                           9841 non-null   float64
 21   ERC20 total ether sent                               9841 non-null   float64
 22   ERC20 total Ether sent contract                      9841 non-null   float64
 23   ERC20 uniq sent addr                                 9841 non-null   float64
 24   ERC20 uniq rec addr                                  9841 non-null   float64
 25   ERC20 uniq rec contract addr                         9841 non-null   float64
 26   ERC20 min val rec                                    9841 non-null   float64
 27   ERC20 max val rec                                    9841 non-null   float64
 28   ERC20 avg val rec                                    9841 non-null   float64
 29   ERC20 min val sent                                   9841 non-null   float64
 30   ERC20 max val sent                                   9841 non-null   float64
 31   ERC20 avg val sent                                   9841 non-null   float64
 32   ERC20 uniq sent token name                           9841 non-null   float64
 33   ERC20 uniq rec token name                            9841 non-null   float64
 34  ERC20_most_sent_token_valid_name                      9841 non-null   int64  
 35  ERC20_most_rec_token_valid_name                       9841 non-null   int64  
dtypes: float64(27), int64(9)
memory usage: 2.7 MB

# Descriptive statistics for all numerical variables
numerical_variables = dataset.select_dtypes(include=['int64', 'float64'])
numerical_variables = numerical_variables.columns
descriptive_stats = dataset[numerical_variables].describe().loc[['mean', 'std', 'min','50%','max']].transpose()
descriptive_stats.columns = ['mean', 'std','min', 'median','max']
descriptive_stats['variance'] = descriptive_stats['std']**2
descriptive_stats['mode'] = dataset[numerical_variables].mode().transpose()[0]
descriptive_stats['mad'] = [stats.median_abs_deviation(dataset[x]) for x in numerical_variables]
descriptive_stats['kurtosis'] = [stats.kurtosis(dataset[x]) for x in numerical_variables]
descriptive_stats['skewness'] = [stats.skew(dataset[x]) for x in numerical_variables]

descriptive_stats

print("Col name", "Number of non-0s")
cols_with_0_mad = descriptive_stats[descriptive_stats['mad'] == 0].index
for x in cols_with_0_mad:
  print(x,dataset[dataset[x] != 0].shape[0])

Col name Number of non-0s
Y 2179
Number of Created Contracts 1356
 Total ERC20 tnxs 4613
 ERC20 total Ether received 4572
 ERC20 total ether sent 1566
 ERC20 total Ether sent contract 28
 ERC20 uniq sent addr 1566
 ERC20 uniq rec addr 4573
 ERC20 uniq rec contract addr 4573
 ERC20 min val rec 2275
 ERC20 max val rec 4425
 ERC20 avg val rec 4420
 ERC20 min val sent 767
 ERC20 max val sent 1477
 ERC20 avg val sent 1468
 ERC20 uniq sent token name 1566
 ERC20 uniq rec token name 4573
ERC20_most_sent_token_valid_name 5441
ERC20_most_rec_token_valid_name 5442

dataset[dataset[' ERC20 total Ether sent contract'] > 0][["Y"]].value_counts()

Y
0    24
1     4
dtype: int64

dataset.drop(' ERC20 total Ether sent contract', axis = 1, inplace = True)

# Plot graph to show the fraud distribution
pie, ax = plt.subplots(figsize=[15, 10])
labels = ['Non-fraud', 'Fraud']
plt.pie(x = dataset['Y'].value_counts(), autopct='%.2f%%', labels=labels)
plt.title('Fraud distribution')
plt.show()

# Select only the specified columns above
num_vs = dataset.select_dtypes(include=['int64', 'float64'])

# Drop last two binary columns
num_vs.drop(columns=['ERC20_most_sent_token_valid_name', 'ERC20_most_rec_token_valid_name'], axis=1, inplace=True)
correlation = num_vs.corr()

# Apply a color gradient, map green background and set 2 decimal places
correlation_styled = correlation.style.background_gradient(cmap='Greens').format("{:.2f}")
correlation_styled

num_vs = dataset.select_dtypes(include=['int64', 'float64'])

numerical_dataset = dataset[list(num_vs)]

def get_redundant_pairs(df):
    # Get diagonal and lower triangular pairs of correlation matrix
    pairs_to_drop = set()
    cols = df.columns
    for i in range(0, df.shape[1]):
        for j in range(0, i+1):
            pairs_to_drop.add((cols[i], cols[j]))
    return pairs_to_drop

def get_top_abs_correlations(df, n=5):
    au_corr = df.corr().abs().unstack()
    labels_to_drop = get_redundant_pairs(df)
    au_corr = au_corr.drop(labels=labels_to_drop).sort_values(ascending=False)
    return au_corr[0:n]

print("Top Absolute Correlations")
print(get_top_abs_correlations(numerical_dataset, 25))

Top Absolute Correlations
 ERC20 total Ether received        ERC20 max val rec                                      0.999967
 ERC20 max val sent                ERC20 avg val sent                                     0.999952
ERC20_most_sent_token_valid_name  ERC20_most_rec_token_valid_name                         0.999794
 ERC20 min val sent                ERC20 avg val sent                                     0.999785
                                   ERC20 max val sent                                     0.999729
 ERC20 total ether sent            ERC20 max val sent                                     0.999649
 ERC20 uniq rec contract addr      ERC20 uniq rec token name                              0.999643
 ERC20 total ether sent            ERC20 avg val sent                                     0.999566
                                   ERC20 min val sent                                     0.999311
 ERC20 total Ether received        ERC20 avg val rec                                      0.859823
 ERC20 max val rec                 ERC20 avg val rec                                      0.859766
Received Tnx                      total transactions (including tnx to create contract    0.806393
 ERC20 uniq sent token name        ERC20 uniq rec token name                              0.789226
 ERC20 uniq rec contract addr      ERC20 uniq sent token name                             0.787966
total Ether sent                  total ether received                                    0.774965
Sent tnx                          total transactions (including tnx to create contract    0.731503
 Total ERC20 tnxs                  ERC20 uniq sent addr                                   0.725989
                                   ERC20 uniq rec addr                                    0.717329
Sent tnx                          Unique Sent To Addresses                                0.670014
Received Tnx                      Unique Received From Addresses                          0.648655
max value received                avg val received                                        0.622959
min val sent                      avg val sent                                            0.594868
 ERC20 uniq sent addr              ERC20 uniq rec addr                                    0.566484
max val sent                      total ether balance                                     0.564872
Unique Received From Addresses    total transactions (including tnx to create contract    0.523848
dtype: float64

perfect_corr_cols = [' ERC20 max val sent', ' ERC20 min val sent', ' ERC20 avg val sent', ' ERC20 max val rec', ' ERC20 uniq rec contract addr']
# Re-check corr. matrix
# Drop cols w perfect corr
num_vs.drop(columns=perfect_corr_cols, axis=1, inplace=True)
correlation_new = num_vs.corr()

# Apply a color gradient, map green background and set 2 decimal places
correlation_styled = correlation_new.style.background_gradient(cmap='Greens').format("{:.2f}")
correlation_styled

highly_corr_cols = [ 'total transactions (including tnx to create contract', 'Unique Received From Addresses', 'Unique Sent To Addresses', ' Total ERC20 tnxs', ' ERC20 avg val rec', 'avg val received']

# Re-check corr. matrix
# Drop cols w perfect corr
num_vs.drop(columns=highly_corr_cols, axis=1, inplace=True)
correlation_new1 = num_vs.corr()

# Apply a color gradient, map green background and set 2 decimal places
correlation_styled = correlation_new1.style.background_gradient(cmap='Greens').format("{:.2f}")
correlation_styled

# Corr. Matrix looks ok now, apply column deletion on actual dataset
drop_columns_corr = perfect_corr_cols + highly_corr_cols
dataset.drop(columns=drop_columns_corr, axis=1, inplace=True)

dataset.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 9841 entries, 0 to 9840
Data columns (total 24 columns):
 #   Column                                   Non-Null Count  Dtype  
---  ------                                   --------------  -----  
 0   Y                                        9841 non-null   int64  
 1   Avg min between sent tnx                 9841 non-null   float64
 2   Avg min between received tnx             9841 non-null   float64
 3   Time Diff between first and last (Mins)  9841 non-null   float64
 4   Sent tnx                                 9841 non-null   int64  
 5   Received Tnx                             9841 non-null   int64  
 6   Number of Created Contracts              9841 non-null   int64  
 7   min value received                       9841 non-null   float64
 8   max value received                       9841 non-null   float64
 9   min val sent                             9841 non-null   float64
 10  max val sent                             9841 non-null   float64
 11  avg val sent                             9841 non-null   float64
 12  total Ether sent                         9841 non-null   float64
 13  total ether received                     9841 non-null   float64
 14  total ether balance                      9841 non-null   float64
 15   ERC20 total Ether received              9841 non-null   float64
 16   ERC20 total ether sent                  9841 non-null   float64
 17   ERC20 uniq sent addr                    9841 non-null   float64
 18   ERC20 uniq rec addr                     9841 non-null   float64
 19   ERC20 min val rec                       9841 non-null   float64
 20   ERC20 uniq sent token name              9841 non-null   float64
 21   ERC20 uniq rec token name               9841 non-null   float64
 22  ERC20_most_sent_token_valid_name         9841 non-null   int64  
 23  ERC20_most_rec_token_valid_name          9841 non-null   int64  
dtypes: float64(18), int64(6)
memory usage: 1.8 MB

numeric_columns = dataset.dtypes[dataset.dtypes != 'object'].index
numeric_columns = list(numeric_columns)
numeric_columns.remove('Y')
# Determine the number of rows and columns for subplots
num_rows = (len(numeric_columns) + 3) // 4
num_cols = 4
# Creating box plots for each numeric column based on 'Response'
plt.figure(figsize=(15, 5*num_rows))
for idx, column in enumerate(numeric_columns, start=1):
    plt.subplot(num_rows, num_cols, idx)
    sns.boxplot(data=dataset, x='Y', y=column)
    plt.title(column)
plt.tight_layout()
plt.show()

features_to_plot = ["Time Diff between first and last (Mins)", "avg val sent", "Sent tnx", "Received Tnx", " ERC20 uniq sent addr", " ERC20 uniq rec addr", "Y"]

# Print a title for the plot.
print("Relative Plot of Selected Features: A Data Subset")

# Create a pair plot.
sns.pairplot(dataset[features_to_plot], hue="Y")

# Display the plot.
plt.show()

Relative Plot of Selected Features: A Data Subset

# Calculate the Sent-to-Received Ratio, replacing infinity with NaN
dataset['Sent_to_Received_Ratio'] = dataset['Sent tnx'] / dataset['Received Tnx']
dataset['Sent_to_Received_Ratio'].fillna(1, inplace=True)
dataset['Sent_to_Received_Ratio'].replace([np.inf, -np.inf], 1, inplace=True)

# Boxplot with outliers
sns.boxplot(data=dataset, x='Y', y='Sent_to_Received_Ratio')

<Axes: xlabel='Y', ylabel='Sent_to_Received_Ratio'>

# Boxplot without outliers
sns.boxplot(data=dataset, x='Y', y='Sent_to_Received_Ratio', showfliers = False)

<Axes: xlabel='Y', ylabel='Sent_to_Received_Ratio'>

# Calculate the Ether_Sent_to_Received_Ratio Ratio, same logic as sent-to-receive ratio above, impute those dividing by 0 by 1
dataset['Ether_Sent_to_Received_Ratio'] = dataset['total Ether sent'] / dataset['total ether received']
dataset['Ether_Sent_to_Received_Ratio'].fillna(1, inplace=True)
dataset['Ether_Sent_to_Received_Ratio'].replace([np.inf, -np.inf], 1, inplace=True)

# Box plot with outliers
sns.boxplot(data=dataset, x='Y', y='Ether_Sent_to_Received_Ratio')

<Axes: xlabel='Y', ylabel='Ether_Sent_to_Received_Ratio'>

# Boxplot without outliers
sns.boxplot(data=dataset, x='Y', y='Ether_Sent_to_Received_Ratio', showfliers = False)

<Axes: xlabel='Y', ylabel='Ether_Sent_to_Received_Ratio'>

# Drop the original columns
dataset.drop(columns = ['total Ether sent', 'total ether received', 'Sent tnx', 'Received Tnx'], inplace=True, axis=1)

# Check that the columns are dropped correctly
dataset.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 9841 entries, 0 to 9840
Data columns (total 22 columns):
 #   Column                                   Non-Null Count  Dtype  
---  ------                                   --------------  -----  
 0   Y                                        9841 non-null   int64  
 1   Avg min between sent tnx                 9841 non-null   float64
 2   Avg min between received tnx             9841 non-null   float64
 3   Time Diff between first and last (Mins)  9841 non-null   float64
 4   Number of Created Contracts              9841 non-null   int64  
 5   min value received                       9841 non-null   float64
 6   max value received                       9841 non-null   float64
 7   min val sent                             9841 non-null   float64
 8   max val sent                             9841 non-null   float64
 9   avg val sent                             9841 non-null   float64
 10  total ether balance                      9841 non-null   float64
 11   ERC20 total Ether received              9841 non-null   float64
 12   ERC20 total ether sent                  9841 non-null   float64
 13   ERC20 uniq sent addr                    9841 non-null   float64
 14   ERC20 uniq rec addr                     9841 non-null   float64
 15   ERC20 min val rec                       9841 non-null   float64
 16   ERC20 uniq sent token name              9841 non-null   float64
 17   ERC20 uniq rec token name               9841 non-null   float64
 18  ERC20_most_sent_token_valid_name         9841 non-null   int64  
 19  ERC20_most_rec_token_valid_name          9841 non-null   int64  
 20  Sent_to_Received_Ratio                   9841 non-null   float64
 21  Ether_Sent_to_Received_Ratio             9841 non-null   float64
dtypes: float64(18), int64(4)
memory usage: 1.7 MB

numeric_columns = dataset.dtypes[dataset.dtypes != 'object'].index
numeric_columns = list(numeric_columns)
numeric_columns.remove('Y')
# Determine the number of rows and columns for subplots
num_rows = (len(numeric_columns) + 3) // 4
num_cols = 4
# Creating box plots for each numeric column based on 'Response'
plt.figure(figsize=(20, 5*num_rows))
for idx, column in enumerate(numeric_columns, start=1):
    plt.subplot(num_rows, num_cols, idx)
    plt.hist(dataset[column])
    plt.title(column + " skewness: " +str(round(stats.skew(dataset[column]),2)))
plt.tight_layout()
plt.show()

# Applying log transformation on the variables
numeric_columns = dataset.dtypes[dataset.dtypes != 'object'].index
numeric_columns = list(numeric_columns)
numeric_columns.remove('Y')
numeric_columns.remove('ERC20_most_sent_token_valid_name')
numeric_columns.remove('ERC20_most_rec_token_valid_name')
# Remove total ether balance because variable is already somewhat normal
numeric_columns.remove('total ether balance')
# Determine the number of rows and columns for subplots
num_rows = (len(numeric_columns) + 3) // 4
num_cols = 4
# Creating box plots for each numeric column based on 'Response'
plt.figure(figsize=(20, 5*num_rows))
for idx, column in enumerate(numeric_columns, start=1):
    plt.subplot(num_rows, num_cols, idx)
    log_transformed = np.log1p(dataset[column])
    plt.hist(log_transformed)
    plt.title("log " + column + " skewness: " +str(round(stats.skew(log_transformed),2)))
plt.tight_layout()
plt.show()

dataset_log = dataset.copy(deep = True)

numeric_columns

['Avg min between sent tnx',
 'Avg min between received tnx',
 'Time Diff between first and last (Mins)',
 'Number of Created Contracts',
 'min value received',
 'max value received ',
 'min val sent',
 'max val sent',
 'avg val sent',
 ' ERC20 total Ether received',
 ' ERC20 total ether sent',
 ' ERC20 uniq sent addr',
 ' ERC20 uniq rec addr',
 ' ERC20 min val rec',
 ' ERC20 uniq sent token name',
 ' ERC20 uniq rec token name',
 'Sent_to_Received_Ratio',
 'Ether_Sent_to_Received_Ratio']

for x in numeric_columns:
    dataset_log[x] = np.log1p(dataset_log[x])

dataset_log

y = dataset_log.Y
X = dataset_log.drop('Y', axis = 1)

train_ratio = 0.70
validation_ratio = 0.20
test_ratio = 0.10
# Taken form :https://datascience.stackexchange.com/questions/15135/train-test-validation-set-splitting-in-sklearn


X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=1 - train_ratio, random_state = 42, stratify = y)

X_val, X_test, y_val, y_test = train_test_split(X_test, y_test, test_size=test_ratio/(test_ratio + validation_ratio),random_state = 1,stratify = y_test)

# Check rows
print("training set shape: ", X_train.shape)
print("validation set shape: ", X_val.shape)
print("testing set shape: ", X_test.shape)

training set shape:  (6888, 21)
validation set shape:  (1968, 21)
testing set shape:  (985, 21)

# Append total ether balance back after log trf step
numeric_columns.append('total ether balance')
scaler = StandardScaler()
X_train[numeric_columns] = scaler.fit_transform(X_train[numeric_columns])
X_val[numeric_columns] = scaler.transform(X_val[numeric_columns])
X_test[numeric_columns] = scaler.transform(X_test[numeric_columns])

training = X_train.copy(deep = True)
training['Class'] = y_train
plt.figure(figsize=(12,28*4))
gs = gridspec.GridSpec(28, 1)
for i, cn in enumerate(training.columns):
    ax = plt.subplot(gs[i])
    sns.distplot(training[cn][training.Class == 1], bins=50)
    sns.distplot(training[cn][training.Class == 0], bins=50)
    ax.set_xlabel('')
    ax.set_title('histogram of feature: ' + str(cn))
plt.show()

X_train.isnull().sum()

Avg min between sent tnx                   0
Avg min between received tnx               0
Time Diff between first and last (Mins)    0
Number of Created Contracts                0
min value received                         0
max value received                         0
min val sent                               0
max val sent                               0
avg val sent                               0
total ether balance                        0
 ERC20 total Ether received                0
 ERC20 total ether sent                    0
 ERC20 uniq sent addr                      0
 ERC20 uniq rec addr                       0
 ERC20 min val rec                         0
 ERC20 uniq sent token name                0
 ERC20 uniq rec token name                 0
ERC20_most_sent_token_valid_name           0
ERC20_most_rec_token_valid_name            0
Sent_to_Received_Ratio                     0
Ether_Sent_to_Received_Ratio               0
dtype: int64

sm = SMOTE(random_state = 42)
X_train_res, y_train_res = sm.fit_resample(X_train, y_train)

print("rows of fraud transaction in training set:" , y_train_res[y_train_res == 1].shape)
print("rows of non-fraud transaction in training set:" , y_train_res[y_train_res == 0].shape)

rows of fraud transaction in training set: (5363,)
rows of non-fraud transaction in training set: (5363,)

features_to_plot = ["max value received ", "min value received", "Y", "max val sent", "min val sent"]

# Print a title for the plot.
print("Relative Plot of Selected Features: A Data Subset")

# Create a pair plot.
sns.pairplot(dataset[features_to_plot], hue="Y")

# Display the plot.
plt.show()

Relative Plot of Selected Features: A Data Subset

# Function to evaluate the models, print confusion matrics if evaluated on test set
def evaluate(model, model_name, evaluated_set, test_features, test_labels):
    y_pred = model.predict(test_features)
    accuracy = accuracy_score(test_labels, y_pred)
    precision = precision_score(test_labels, y_pred)
    recall = recall_score(test_labels, y_pred)
    f1 = f1_score(test_labels, y_pred)
    roc_auc = roc_auc_score(test_labels, y_pred)
    conf_matrix = confusion_matrix(test_labels, y_pred)

    print("Model evaluation for: " + model_name + " on " + evaluated_set)
    print(f"Accuracy: {accuracy}")
    print(f"Precision: {precision}")
    print(f"Recall: {recall}")
    print(f"F1 Score: {f1}")
    print(f"ROC AUC Score: {roc_auc}")
    print(f"Confusion Matrix:\n{conf_matrix}")

    if (evaluated_set == "test set"):
      labels = [0, 1]
      cm = confusion_matrix(test_labels, y_pred, labels = labels)

      fig = plt.figure(figsize=(16,8))

      plt1 = ConfusionMatrixDisplay(cm, display_labels =labels)
      plt1.plot()

    return accuracy

# Actual test set confusion metrics for reference
labels = [0, 1]
actual_cm = confusion_matrix(y_test, y_test, labels = labels)
plt2 = ConfusionMatrixDisplay(confusion_matrix= actual_cm, display_labels=labels)
plt2.plot()

<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay at 0x78f248f02f80>

svc = SVC(random_state=42)
svc.fit(X_train_res,y_train_res)
evaluate(svc, "Base Model", "validation set",X_val, y_val )

Model evaluation for: Base Model on validation set
Accuracy: 0.9563008130081301
Precision: 0.8723404255319149
Recall: 0.9403669724770642
F1 Score: 0.9050772626931567
ROC AUC Score: 0.9506012408077227
Confusion Matrix:
[[1472   60]
 [  26  410]]

0.9563008130081301

# grid = {
#     'C':[0.01,0.1,1,10],
#     'kernel' : ["linear","poly","rbf","sigmoid"],
#     'degree' : [1,2,3,4,5,6,7],
#     'gamma' : [0.01,1]
# }

# svc = SVC(random_state=42)

# svm_cv = GridSearchCV(svc, grid, cv=5)
# svm_cv.fit(X_train,y_train)

# print("Best Parameters:",svm_cv.best_params_)

# Running this code takes a very long time (~45minutes) so the results but the results are shown in the text block below.

# Use the best parameters and compare performance
svc = SVC(random_state=42, C=0.01, degree=3, gamma=1, kernel='poly')
svc.fit(X_train_res,y_train_res)
evaluate(svc, "Tuned Model", "validation set",X_val, y_val )
print("\n")
evaluate(svc, "Tuned Model", "test set", X_test, y_test )

Model evaluation for: Tuned Model on validation set
Accuracy: 0.9659552845528455
Precision: 0.9002169197396963
Recall: 0.9518348623853211
F1 Score: 0.9253065774804905
ROC AUC Score: 0.9609043763623734
Confusion Matrix:
[[1486   46]
 [  21  415]]


Model evaluation for: Tuned Model on test set
Accuracy: 0.9573604060913705
Precision: 0.8893805309734514
Recall: 0.9220183486238532
F1 Score: 0.9054054054054055
ROC AUC Score: 0.9447119122519527
Confusion Matrix:
[[742  25]
 [ 17 201]]

0.9573604060913705

<Figure size 1600x800 with 0 Axes>

rfc = RandomForestClassifier(random_state = 42)
print('Parameters currently in use:\n')
print(rfc.get_params())

Parameters currently in use:

{'bootstrap': True, 'ccp_alpha': 0.0, 'class_weight': None, 'criterion': 'gini', 'max_depth': None, 'max_features': 'sqrt', 'max_leaf_nodes': None, 'max_samples': None, 'min_impurity_decrease': 0.0, 'min_samples_leaf': 1, 'min_samples_split': 2, 'min_weight_fraction_leaf': 0.0, 'n_estimators': 100, 'n_jobs': None, 'oob_score': False, 'random_state': 42, 'verbose': 0, 'warm_start': False}

rfc.fit(X_train_res, y_train_res)
evaluate(rfc, "Base Model", "validation set", X_val, y_val)

Model evaluation for: Base Model on validation set
Accuracy: 0.9761178861788617
Precision: 0.9471264367816092
Recall: 0.944954128440367
F1 Score: 0.9460390355912743
ROC AUC Score: 0.9649705368050399
Confusion Matrix:
[[1509   23]
 [  24  412]]

0.9761178861788617

n_estimators = [int(x) for x in np.linspace(start = 200, stop = 2000, num = 10)]
max_features = ['auto', 'sqrt']
max_depth = [int(x) for x in np.linspace(10, 110, num = 11)]
max_depth.append(None)
min_samples_split = [2, 5, 10]
min_samples_leaf = [1, 2, 4]
bootstrap = [True]
random_grid = {'n_estimators': n_estimators,
               'max_features': max_features,
               'max_depth': max_depth,
               'min_samples_split': min_samples_split,
               'min_samples_leaf': min_samples_leaf,
               'bootstrap': bootstrap}
print(random_grid)

{'n_estimators': [200, 400, 600, 800, 1000, 1200, 1400, 1600, 1800, 2000], 'max_features': ['auto', 'sqrt'], 'max_depth': [10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, None], 'min_samples_split': [2, 5, 10], 'min_samples_leaf': [1, 2, 4], 'bootstrap': [True]}

warnings.simplefilter(action='ignore', category=FutureWarning)

model = RandomForestClassifier(random_state = 42)
model_random = RandomizedSearchCV(estimator = model, param_distributions = random_grid, n_iter = 100, cv = 3, verbose = 2, n_jobs = -1, random_state = 42)
model_random.fit(X_train_res, y_train_res)

Fitting 3 folds for each of 100 candidates, totalling 300 fits

RandomizedSearchCV(cv=3, estimator=RandomForestClassifier(random_state=42),
                   n_iter=100, n_jobs=-1,
                   param_distributions={'bootstrap': [True],
                                        'max_depth': [10, 20, 30, 40, 50, 60,
                                                      70, 80, 90, 100, 110,
                                                      None],
                                        'max_features': ['auto', 'sqrt'],
                                        'min_samples_leaf': [1, 2, 4],
                                        'min_samples_split': [2, 5, 10],
                                        'n_estimators': [200, 400, 600, 800,
                                                         1000, 1200, 1400, 1600,
                                                         1800, 2000]},
                   random_state=42, verbose=2)

RandomizedSearchCV(cv=3, estimator=RandomForestClassifier(random_state=42),
                   n_iter=100, n_jobs=-1,
                   param_distributions={'bootstrap': [True],
                                        'max_depth': [10, 20, 30, 40, 50, 60,
                                                      70, 80, 90, 100, 110,
                                                      None],
                                        'max_features': ['auto', 'sqrt'],
                                        'min_samples_leaf': [1, 2, 4],
                                        'min_samples_split': [2, 5, 10],
                                        'n_estimators': [200, 400, 600, 800,
                                                         1000, 1200, 1400, 1600,
                                                         1800, 2000]},
                   random_state=42, verbose=2)

RandomForestClassifier(random_state=42)

RandomForestClassifier(random_state=42)

model_random.best_params_

{'n_estimators': 2000,
 'min_samples_split': 2,
 'min_samples_leaf': 1,
 'max_features': 'sqrt',
 'max_depth': None,
 'bootstrap': True}

# Best_random = model_random.best_estimator_
best_random = RandomForestClassifier(n_estimators = 2000, min_samples_split = 2, min_samples_leaf = 1, max_features = "sqrt", max_depth = None, bootstrap = True, random_state = 42)
best_random.fit(X_train_res, y_train_res)
random_accuracy = evaluate(best_random, "Best Random Model", "validation set", X_val, y_val)

Model evaluation for: Best Random Model on validation set
Accuracy: 0.9796747967479674
Precision: 0.952054794520548
Recall: 0.9564220183486238
F1 Score: 0.954233409610984
ROC AUC Score: 0.9713572232735286
Confusion Matrix:
[[1511   21]
 [  19  417]]

param_grid = {
    'n_estimators': [1900, 2000, 2100],
    'max_depth': [None,1,2],
    'max_features' : ['auto', 'sqrt'],
    'min_samples_split': [2, 3, 4],
    'min_samples_leaf': [1, 2],
    'bootstrap': [True],
}

model_grid = RandomForestClassifier(random_state = 42)
grid_search = GridSearchCV(estimator = model_grid, param_grid = param_grid,
                          cv = 3, verbose = 2, n_jobs = -1)

grid_search.fit(X_train_res, y_train_res)

grid_search.best_params_

Fitting 3 folds for each of 108 candidates, totalling 324 fits

{'bootstrap': True,
 'max_depth': None,
 'max_features': 'auto',
 'min_samples_leaf': 1,
 'min_samples_split': 2,
 'n_estimators': 2000}

rfc = RandomForestClassifier(n_estimators = 2000, min_samples_split = 2, min_samples_leaf = 1, max_features = 'sqrt',max_depth = None, bootstrap = True,  random_state = 42)
rfc.fit(X_train_res, y_train_res)
grid_accuracy = evaluate(rfc, "Tuned Model", "test set", X_test, y_test)

Model evaluation for: Tuned Model on test set
Accuracy: 0.9736040609137055
Precision: 0.9571428571428572
Recall: 0.9220183486238532
F1 Score: 0.9392523364485982
ROC AUC Score: 0.9551421599703361
Confusion Matrix:
[[758   9]
 [ 17 201]]

<Figure size 1600x800 with 0 Axes>

# Reshape data into 1D format for a 1D CNN
X_train_res = np.expand_dims(X_train_res, axis=2)
X_val = np.expand_dims(X_val, axis=2)
X_test = np.expand_dims(X_test, axis=2)

# Define the 1D CNN model
model = tf.keras.Sequential([
    tf.keras.layers.Conv1D(64, 3, activation='relu', input_shape=(X_train_res.shape[1], 1)),
    tf.keras.layers.MaxPooling1D(2),
    tf.keras.layers.Flatten(),
    tf.keras.layers.Dense(128, activation='relu'),
    tf.keras.layers.Dropout(0.5),
    tf.keras.layers.Dense(1, activation='sigmoid')
])

# Compile the model
model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])

history = model.fit(
            X_train_res,
            y_train_res,
            epochs=20,
            batch_size=300,
            validation_data=(X_val, y_val),
            verbose=2
        )

# Evaluate the model on the validation set
val_accuracy = history.history['val_accuracy'][-1]
print(f"Validation accuracy for pre-tuned model: {val_accuracy}")


# Evaluate the model on the test set
y_pred = model.predict(X_test)
y_pred_binary = (y_pred > 0.5).astype(int)  # Convert probabilities to binary predictions (0 or 1)

accuracy = accuracy_score(y_test, y_pred_binary)
precision = precision_score(y_test, y_pred_binary)
recall = recall_score(y_test, y_pred_binary)
f1 = f1_score(y_test, y_pred_binary)
roc_auc = roc_auc_score(y_test, y_pred_binary)
conf_matrix = confusion_matrix(y_test, y_pred_binary)

print(f"Accuracy: {accuracy}")
print(f"Precision: {precision}")
print(f"Recall: {recall}")
print(f"F1 Score: {f1}")
print(f"ROC AUC Score: {roc_auc}")
print(f"Confusion Matrix:\n{conf_matrix}")

Epoch 1/20
36/36 - 2s - loss: 0.4112 - accuracy: 0.8379 - val_loss: 0.2497 - val_accuracy: 0.9187 - 2s/epoch - 65ms/step
Epoch 2/20
36/36 - 1s - loss: 0.2140 - accuracy: 0.9241 - val_loss: 0.1975 - val_accuracy: 0.9299 - 628ms/epoch - 17ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1673 - accuracy: 0.9405 - val_loss: 0.1431 - val_accuracy: 0.9487 - 617ms/epoch - 17ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1490 - accuracy: 0.9450 - val_loss: 0.1474 - val_accuracy: 0.9461 - 644ms/epoch - 18ms/step
Epoch 5/20
36/36 - 1s - loss: 0.1373 - accuracy: 0.9501 - val_loss: 0.1393 - val_accuracy: 0.9492 - 603ms/epoch - 17ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1258 - accuracy: 0.9534 - val_loss: 0.1265 - val_accuracy: 0.9517 - 424ms/epoch - 12ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1221 - accuracy: 0.9542 - val_loss: 0.1247 - val_accuracy: 0.9512 - 423ms/epoch - 12ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1133 - accuracy: 0.9562 - val_loss: 0.1113 - val_accuracy: 0.9548 - 435ms/epoch - 12ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1093 - accuracy: 0.9586 - val_loss: 0.1142 - val_accuracy: 0.9533 - 396ms/epoch - 11ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1051 - accuracy: 0.9593 - val_loss: 0.1157 - val_accuracy: 0.9533 - 431ms/epoch - 12ms/step
Epoch 11/20
36/36 - 0s - loss: 0.1024 - accuracy: 0.9597 - val_loss: 0.1217 - val_accuracy: 0.9492 - 389ms/epoch - 11ms/step
Epoch 12/20
36/36 - 0s - loss: 0.0987 - accuracy: 0.9596 - val_loss: 0.1178 - val_accuracy: 0.9522 - 390ms/epoch - 11ms/step
Epoch 13/20
36/36 - 0s - loss: 0.0969 - accuracy: 0.9606 - val_loss: 0.1059 - val_accuracy: 0.9578 - 398ms/epoch - 11ms/step
Epoch 14/20
36/36 - 0s - loss: 0.0938 - accuracy: 0.9621 - val_loss: 0.0973 - val_accuracy: 0.9593 - 390ms/epoch - 11ms/step
Epoch 15/20
36/36 - 0s - loss: 0.0931 - accuracy: 0.9621 - val_loss: 0.0999 - val_accuracy: 0.9619 - 435ms/epoch - 12ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0893 - accuracy: 0.9647 - val_loss: 0.1098 - val_accuracy: 0.9548 - 402ms/epoch - 11ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0842 - accuracy: 0.9661 - val_loss: 0.0942 - val_accuracy: 0.9619 - 422ms/epoch - 12ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0836 - accuracy: 0.9663 - val_loss: 0.0978 - val_accuracy: 0.9599 - 427ms/epoch - 12ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0832 - accuracy: 0.9664 - val_loss: 0.0998 - val_accuracy: 0.9604 - 383ms/epoch - 11ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0822 - accuracy: 0.9661 - val_loss: 0.0889 - val_accuracy: 0.9654 - 385ms/epoch - 11ms/step
Validation accuracy for pre-tuned model: 0.9654471278190613
31/31 [==============================] - 0s 2ms/step
Accuracy: 0.9614213197969543
Precision: 0.9090909090909091
Recall: 0.9174311926605505
F1 Score: 0.9132420091324202
ROC AUC Score: 0.9456777866822961
Confusion Matrix:
[[747  20]
 [ 18 200]]

# Hyperparameter tuning
learning_rates = [0.001, 0.01, 0.1]
num_filters = [32, 64, 128]
dropout_rate = [0.3, 0.5, 0.7]
filter_size = [3, 5, 7]


best_val_accuracy = 0
best_model = None
best_model_parameters = {}
num_epochs = 20
batch_size = 300


for lr in learning_rates:
  for filter in num_filters:
    for dr in dropout_rate:
      for size in filter_size:
        model = tf.keras.Sequential([
            tf.keras.layers.Conv1D(filter, size, activation='relu', input_shape=(X_train_res.shape[1], 1)),
            tf.keras.layers.MaxPooling1D(2),
            tf.keras.layers.Flatten(),
            tf.keras.layers.Dense(128, activation='relu'),
            tf.keras.layers.Dropout(dr),
            tf.keras.layers.Dense(1, activation='sigmoid')
        ])
        model.compile(optimizer=tf.keras.optimizers.Adam(learning_rate=lr), loss='binary_crossentropy', metrics=['accuracy'])

        # Train the model on the training data
        history = model.fit(
            X_train_res,
            y_train_res,
            epochs=num_epochs,
            batch_size=batch_size,
            validation_data=(X_val, y_val),
            verbose=2
        )

        # Evaluate the model on the validation set
        val_accuracy = history.history['val_accuracy'][-1]
        print(f"Validation accuracy for Model of [Learning Rate {lr} | Num Filters {filter} | Dropout Rate {dr} | FilterSize {size}]: {val_accuracy}")

        if val_accuracy > best_val_accuracy:
            best_val_accuracy = val_accuracy
            best_model = model
            best_model_parameters = {
                "learning_rate": lr,
                "num_filters": filter,
                "droupout_rate": dr,
                "filter_size": size
            }

Epoch 1/20
36/36 - 1s - loss: 0.4271 - accuracy: 0.8409 - val_loss: 0.2740 - val_accuracy: 0.9167 - 1s/epoch - 35ms/step
Epoch 2/20
36/36 - 0s - loss: 0.2321 - accuracy: 0.9177 - val_loss: 0.1943 - val_accuracy: 0.9350 - 290ms/epoch - 8ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1792 - accuracy: 0.9397 - val_loss: 0.1572 - val_accuracy: 0.9446 - 276ms/epoch - 8ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1554 - accuracy: 0.9464 - val_loss: 0.1325 - val_accuracy: 0.9543 - 292ms/epoch - 8ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1404 - accuracy: 0.9502 - val_loss: 0.1385 - val_accuracy: 0.9512 - 275ms/epoch - 8ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1346 - accuracy: 0.9504 - val_loss: 0.1353 - val_accuracy: 0.9512 - 290ms/epoch - 8ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1258 - accuracy: 0.9539 - val_loss: 0.1330 - val_accuracy: 0.9472 - 343ms/epoch - 10ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1201 - accuracy: 0.9549 - val_loss: 0.1312 - val_accuracy: 0.9472 - 454ms/epoch - 13ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1162 - accuracy: 0.9550 - val_loss: 0.1132 - val_accuracy: 0.9548 - 474ms/epoch - 13ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1109 - accuracy: 0.9572 - val_loss: 0.1207 - val_accuracy: 0.9512 - 491ms/epoch - 14ms/step
Epoch 11/20
36/36 - 0s - loss: 0.1029 - accuracy: 0.9594 - val_loss: 0.1123 - val_accuracy: 0.9543 - 487ms/epoch - 14ms/step
Epoch 12/20
36/36 - 0s - loss: 0.1032 - accuracy: 0.9590 - val_loss: 0.1267 - val_accuracy: 0.9482 - 480ms/epoch - 13ms/step
Epoch 13/20
36/36 - 0s - loss: 0.0988 - accuracy: 0.9621 - val_loss: 0.1057 - val_accuracy: 0.9553 - 496ms/epoch - 14ms/step
Epoch 14/20
36/36 - 0s - loss: 0.0951 - accuracy: 0.9622 - val_loss: 0.1055 - val_accuracy: 0.9578 - 453ms/epoch - 13ms/step
Epoch 15/20
36/36 - 0s - loss: 0.0918 - accuracy: 0.9632 - val_loss: 0.1133 - val_accuracy: 0.9533 - 465ms/epoch - 13ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0927 - accuracy: 0.9619 - val_loss: 0.1042 - val_accuracy: 0.9573 - 464ms/epoch - 13ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0878 - accuracy: 0.9642 - val_loss: 0.1167 - val_accuracy: 0.9507 - 424ms/epoch - 12ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0888 - accuracy: 0.9647 - val_loss: 0.1050 - val_accuracy: 0.9583 - 275ms/epoch - 8ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0843 - accuracy: 0.9657 - val_loss: 0.0929 - val_accuracy: 0.9629 - 330ms/epoch - 9ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0830 - accuracy: 0.9663 - val_loss: 0.0931 - val_accuracy: 0.9614 - 290ms/epoch - 8ms/step
Validation accuracy for Model of [Learning Rate 0.001 | Num Filters 32 | Dropout Rate 0.3 | FilterSize 3]: 0.9613820910453796
Epoch 1/20
36/36 - 1s - loss: 0.4660 - accuracy: 0.7946 - val_loss: 0.3023 - val_accuracy: 0.8933 - 1s/epoch - 36ms/step
Epoch 2/20
36/36 - 0s - loss: 0.2607 - accuracy: 0.9057 - val_loss: 0.2183 - val_accuracy: 0.9263 - 291ms/epoch - 8ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1901 - accuracy: 0.9338 - val_loss: 0.1457 - val_accuracy: 0.9466 - 273ms/epoch - 8ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1579 - accuracy: 0.9422 - val_loss: 0.1459 - val_accuracy: 0.9441 - 282ms/epoch - 8ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1375 - accuracy: 0.9497 - val_loss: 0.1308 - val_accuracy: 0.9517 - 275ms/epoch - 8ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1264 - accuracy: 0.9532 - val_loss: 0.1196 - val_accuracy: 0.9548 - 287ms/epoch - 8ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1217 - accuracy: 0.9533 - val_loss: 0.1553 - val_accuracy: 0.9421 - 279ms/epoch - 8ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1114 - accuracy: 0.9566 - val_loss: 0.1207 - val_accuracy: 0.9492 - 278ms/epoch - 8ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1048 - accuracy: 0.9591 - val_loss: 0.1164 - val_accuracy: 0.9507 - 289ms/epoch - 8ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1021 - accuracy: 0.9596 - val_loss: 0.0967 - val_accuracy: 0.9619 - 321ms/epoch - 9ms/step
Epoch 11/20
36/36 - 0s - loss: 0.0987 - accuracy: 0.9601 - val_loss: 0.1023 - val_accuracy: 0.9578 - 284ms/epoch - 8ms/step
Epoch 12/20
36/36 - 0s - loss: 0.0950 - accuracy: 0.9620 - val_loss: 0.1077 - val_accuracy: 0.9568 - 271ms/epoch - 8ms/step
Epoch 13/20
36/36 - 0s - loss: 0.0903 - accuracy: 0.9639 - val_loss: 0.1013 - val_accuracy: 0.9568 - 286ms/epoch - 8ms/step
Epoch 14/20
36/36 - 0s - loss: 0.0879 - accuracy: 0.9644 - val_loss: 0.1053 - val_accuracy: 0.9563 - 280ms/epoch - 8ms/step
Epoch 15/20
36/36 - 0s - loss: 0.0842 - accuracy: 0.9662 - val_loss: 0.0909 - val_accuracy: 0.9649 - 272ms/epoch - 8ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0825 - accuracy: 0.9681 - val_loss: 0.0916 - val_accuracy: 0.9675 - 320ms/epoch - 9ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0822 - accuracy: 0.9672 - val_loss: 0.1256 - val_accuracy: 0.9461 - 283ms/epoch - 8ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0788 - accuracy: 0.9690 - val_loss: 0.1151 - val_accuracy: 0.9507 - 272ms/epoch - 8ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0782 - accuracy: 0.9688 - val_loss: 0.0886 - val_accuracy: 0.9675 - 307ms/epoch - 9ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0776 - accuracy: 0.9693 - val_loss: 0.0956 - val_accuracy: 0.9629 - 438ms/epoch - 12ms/step
Validation accuracy for Model of [Learning Rate 0.001 | Num Filters 32 | Dropout Rate 0.3 | FilterSize 5]: 0.9629064798355103
Epoch 1/20
36/36 - 1s - loss: 0.4879 - accuracy: 0.7863 - val_loss: 0.3615 - val_accuracy: 0.8572 - 1s/epoch - 34ms/step
Epoch 2/20
36/36 - 0s - loss: 0.2843 - accuracy: 0.8993 - val_loss: 0.2261 - val_accuracy: 0.9248 - 316ms/epoch - 9ms/step
Epoch 3/20
36/36 - 0s - loss: 0.2022 - accuracy: 0.9298 - val_loss: 0.1840 - val_accuracy: 0.9324 - 286ms/epoch - 8ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1661 - accuracy: 0.9420 - val_loss: 0.1690 - val_accuracy: 0.9334 - 268ms/epoch - 7ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1469 - accuracy: 0.9480 - val_loss: 0.1342 - val_accuracy: 0.9497 - 260ms/epoch - 7ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1333 - accuracy: 0.9512 - val_loss: 0.1361 - val_accuracy: 0.9502 - 270ms/epoch - 7ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1232 - accuracy: 0.9535 - val_loss: 0.1361 - val_accuracy: 0.9512 - 280ms/epoch - 8ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1155 - accuracy: 0.9571 - val_loss: 0.1167 - val_accuracy: 0.9548 - 316ms/epoch - 9ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1092 - accuracy: 0.9594 - val_loss: 0.1304 - val_accuracy: 0.9472 - 255ms/epoch - 7ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1044 - accuracy: 0.9599 - val_loss: 0.1099 - val_accuracy: 0.9553 - 313ms/epoch - 9ms/step
Epoch 11/20
36/36 - 0s - loss: 0.0995 - accuracy: 0.9605 - val_loss: 0.1175 - val_accuracy: 0.9517 - 258ms/epoch - 7ms/step
Epoch 12/20
36/36 - 0s - loss: 0.0969 - accuracy: 0.9609 - val_loss: 0.0931 - val_accuracy: 0.9644 - 268ms/epoch - 7ms/step
Epoch 13/20
36/36 - 0s - loss: 0.0962 - accuracy: 0.9619 - val_loss: 0.1088 - val_accuracy: 0.9533 - 283ms/epoch - 8ms/step
Epoch 14/20
36/36 - 0s - loss: 0.0897 - accuracy: 0.9651 - val_loss: 0.1081 - val_accuracy: 0.9538 - 290ms/epoch - 8ms/step
Epoch 15/20
36/36 - 0s - loss: 0.0874 - accuracy: 0.9657 - val_loss: 0.1101 - val_accuracy: 0.9512 - 271ms/epoch - 8ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0821 - accuracy: 0.9664 - val_loss: 0.0954 - val_accuracy: 0.9609 - 310ms/epoch - 9ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0789 - accuracy: 0.9670 - val_loss: 0.1042 - val_accuracy: 0.9553 - 277ms/epoch - 8ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0783 - accuracy: 0.9683 - val_loss: 0.0884 - val_accuracy: 0.9654 - 261ms/epoch - 7ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0784 - accuracy: 0.9686 - val_loss: 0.0919 - val_accuracy: 0.9604 - 253ms/epoch - 7ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0731 - accuracy: 0.9705 - val_loss: 0.0931 - val_accuracy: 0.9609 - 258ms/epoch - 7ms/step
Validation accuracy for Model of [Learning Rate 0.001 | Num Filters 32 | Dropout Rate 0.3 | FilterSize 7]: 0.9608739614486694
Epoch 1/20
36/36 - 1s - loss: 0.4754 - accuracy: 0.7887 - val_loss: 0.3060 - val_accuracy: 0.9101 - 1s/epoch - 34ms/step
Epoch 2/20
36/36 - 0s - loss: 0.2609 - accuracy: 0.9047 - val_loss: 0.2105 - val_accuracy: 0.9268 - 276ms/epoch - 8ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1887 - accuracy: 0.9366 - val_loss: 0.1602 - val_accuracy: 0.9436 - 408ms/epoch - 11ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1647 - accuracy: 0.9416 - val_loss: 0.1435 - val_accuracy: 0.9472 - 475ms/epoch - 13ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1498 - accuracy: 0.9480 - val_loss: 0.1542 - val_accuracy: 0.9421 - 426ms/epoch - 12ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1419 - accuracy: 0.9509 - val_loss: 0.1338 - val_accuracy: 0.9512 - 442ms/epoch - 12ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1324 - accuracy: 0.9527 - val_loss: 0.1438 - val_accuracy: 0.9436 - 455ms/epoch - 13ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1277 - accuracy: 0.9529 - val_loss: 0.1172 - val_accuracy: 0.9548 - 430ms/epoch - 12ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1217 - accuracy: 0.9534 - val_loss: 0.1244 - val_accuracy: 0.9512 - 471ms/epoch - 13ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1176 - accuracy: 0.9563 - val_loss: 0.1193 - val_accuracy: 0.9507 - 448ms/epoch - 12ms/step
Epoch 11/20
36/36 - 0s - loss: 0.1125 - accuracy: 0.9569 - val_loss: 0.1108 - val_accuracy: 0.9568 - 471ms/epoch - 13ms/step
Epoch 12/20
36/36 - 0s - loss: 0.1083 - accuracy: 0.9586 - val_loss: 0.1047 - val_accuracy: 0.9599 - 461ms/epoch - 13ms/step
Epoch 13/20
36/36 - 0s - loss: 0.1072 - accuracy: 0.9592 - val_loss: 0.1142 - val_accuracy: 0.9563 - 441ms/epoch - 12ms/step
Epoch 14/20
36/36 - 0s - loss: 0.1039 - accuracy: 0.9596 - val_loss: 0.1005 - val_accuracy: 0.9599 - 261ms/epoch - 7ms/step
Epoch 15/20
36/36 - 0s - loss: 0.0986 - accuracy: 0.9616 - val_loss: 0.1108 - val_accuracy: 0.9558 - 266ms/epoch - 7ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0974 - accuracy: 0.9635 - val_loss: 0.1029 - val_accuracy: 0.9588 - 268ms/epoch - 7ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0944 - accuracy: 0.9630 - val_loss: 0.1086 - val_accuracy: 0.9558 - 287ms/epoch - 8ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0918 - accuracy: 0.9630 - val_loss: 0.1060 - val_accuracy: 0.9573 - 271ms/epoch - 8ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0920 - accuracy: 0.9642 - val_loss: 0.0978 - val_accuracy: 0.9609 - 289ms/epoch - 8ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0906 - accuracy: 0.9631 - val_loss: 0.1054 - val_accuracy: 0.9583 - 330ms/epoch - 9ms/step
Validation accuracy for Model of [Learning Rate 0.001 | Num Filters 32 | Dropout Rate 0.5 | FilterSize 3]: 0.9583333134651184
Epoch 1/20
36/36 - 1s - loss: 0.4788 - accuracy: 0.7969 - val_loss: 0.3236 - val_accuracy: 0.8933 - 1s/epoch - 34ms/step
Epoch 2/20
36/36 - 0s - loss: 0.2768 - accuracy: 0.8990 - val_loss: 0.2203 - val_accuracy: 0.9273 - 278ms/epoch - 8ms/step
Epoch 3/20
36/36 - 0s - loss: 0.2034 - accuracy: 0.9304 - val_loss: 0.1937 - val_accuracy: 0.9314 - 281ms/epoch - 8ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1706 - accuracy: 0.9407 - val_loss: 0.1593 - val_accuracy: 0.9400 - 279ms/epoch - 8ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1499 - accuracy: 0.9459 - val_loss: 0.1306 - val_accuracy: 0.9522 - 277ms/epoch - 8ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1360 - accuracy: 0.9500 - val_loss: 0.1335 - val_accuracy: 0.9487 - 271ms/epoch - 8ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1288 - accuracy: 0.9522 - val_loss: 0.1341 - val_accuracy: 0.9482 - 299ms/epoch - 8ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1212 - accuracy: 0.9536 - val_loss: 0.1144 - val_accuracy: 0.9558 - 272ms/epoch - 8ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1184 - accuracy: 0.9552 - val_loss: 0.1568 - val_accuracy: 0.9416 - 287ms/epoch - 8ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1113 - accuracy: 0.9576 - val_loss: 0.1061 - val_accuracy: 0.9573 - 279ms/epoch - 8ms/step
Epoch 11/20
36/36 - 0s - loss: 0.1095 - accuracy: 0.9575 - val_loss: 0.1103 - val_accuracy: 0.9527 - 338ms/epoch - 9ms/step
Epoch 12/20
36/36 - 0s - loss: 0.1028 - accuracy: 0.9596 - val_loss: 0.1051 - val_accuracy: 0.9578 - 274ms/epoch - 8ms/step
Epoch 13/20
36/36 - 0s - loss: 0.1039 - accuracy: 0.9587 - val_loss: 0.0993 - val_accuracy: 0.9599 - 274ms/epoch - 8ms/step
Epoch 14/20
36/36 - 0s - loss: 0.0974 - accuracy: 0.9615 - val_loss: 0.0980 - val_accuracy: 0.9624 - 285ms/epoch - 8ms/step
Epoch 15/20
36/36 - 0s - loss: 0.0967 - accuracy: 0.9607 - val_loss: 0.1082 - val_accuracy: 0.9578 - 279ms/epoch - 8ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0910 - accuracy: 0.9640 - val_loss: 0.1061 - val_accuracy: 0.9553 - 270ms/epoch - 7ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0913 - accuracy: 0.9619 - val_loss: 0.1034 - val_accuracy: 0.9548 - 275ms/epoch - 8ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0880 - accuracy: 0.9658 - val_loss: 0.1078 - val_accuracy: 0.9527 - 270ms/epoch - 7ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0861 - accuracy: 0.9626 - val_loss: 0.1068 - val_accuracy: 0.9588 - 272ms/epoch - 8ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0852 - accuracy: 0.9661 - val_loss: 0.0962 - val_accuracy: 0.9624 - 266ms/epoch - 7ms/step
Validation accuracy for Model of [Learning Rate 0.001 | Num Filters 32 | Dropout Rate 0.5 | FilterSize 5]: 0.9623983502388
Epoch 1/20
36/36 - 1s - loss: 0.4954 - accuracy: 0.7739 - val_loss: 0.3384 - val_accuracy: 0.8699 - 1s/epoch - 35ms/step
Epoch 2/20
36/36 - 0s - loss: 0.2844 - accuracy: 0.8928 - val_loss: 0.2352 - val_accuracy: 0.9197 - 268ms/epoch - 7ms/step
Epoch 3/20
36/36 - 0s - loss: 0.2056 - accuracy: 0.9260 - val_loss: 0.1822 - val_accuracy: 0.9299 - 269ms/epoch - 7ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1742 - accuracy: 0.9352 - val_loss: 0.1664 - val_accuracy: 0.9360 - 277ms/epoch - 8ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1548 - accuracy: 0.9421 - val_loss: 0.1484 - val_accuracy: 0.9456 - 264ms/epoch - 7ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1394 - accuracy: 0.9475 - val_loss: 0.1333 - val_accuracy: 0.9482 - 258ms/epoch - 7ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1297 - accuracy: 0.9509 - val_loss: 0.1257 - val_accuracy: 0.9527 - 269ms/epoch - 7ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1224 - accuracy: 0.9527 - val_loss: 0.1200 - val_accuracy: 0.9553 - 267ms/epoch - 7ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1143 - accuracy: 0.9567 - val_loss: 0.1261 - val_accuracy: 0.9517 - 273ms/epoch - 8ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1095 - accuracy: 0.9585 - val_loss: 0.1294 - val_accuracy: 0.9492 - 275ms/epoch - 8ms/step
Epoch 11/20
36/36 - 0s - loss: 0.1044 - accuracy: 0.9601 - val_loss: 0.1095 - val_accuracy: 0.9563 - 300ms/epoch - 8ms/step
Epoch 12/20
36/36 - 0s - loss: 0.1024 - accuracy: 0.9611 - val_loss: 0.1166 - val_accuracy: 0.9543 - 281ms/epoch - 8ms/step
Epoch 13/20
36/36 - 0s - loss: 0.0957 - accuracy: 0.9632 - val_loss: 0.1171 - val_accuracy: 0.9543 - 319ms/epoch - 9ms/step
Epoch 14/20
36/36 - 0s - loss: 0.0941 - accuracy: 0.9629 - val_loss: 0.1217 - val_accuracy: 0.9512 - 336ms/epoch - 9ms/step
Epoch 15/20
36/36 - 0s - loss: 0.0914 - accuracy: 0.9649 - val_loss: 0.1184 - val_accuracy: 0.9507 - 284ms/epoch - 8ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0893 - accuracy: 0.9649 - val_loss: 0.1116 - val_accuracy: 0.9543 - 266ms/epoch - 7ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0861 - accuracy: 0.9651 - val_loss: 0.0979 - val_accuracy: 0.9624 - 269ms/epoch - 7ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0844 - accuracy: 0.9673 - val_loss: 0.0945 - val_accuracy: 0.9639 - 271ms/epoch - 8ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0820 - accuracy: 0.9666 - val_loss: 0.1062 - val_accuracy: 0.9568 - 270ms/epoch - 7ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0805 - accuracy: 0.9697 - val_loss: 0.1023 - val_accuracy: 0.9593 - 267ms/epoch - 7ms/step
Validation accuracy for Model of [Learning Rate 0.001 | Num Filters 32 | Dropout Rate 0.5 | FilterSize 7]: 0.9593495726585388
Epoch 1/20
36/36 - 1s - loss: 0.4642 - accuracy: 0.7955 - val_loss: 0.2887 - val_accuracy: 0.9101 - 1s/epoch - 34ms/step
Epoch 2/20
36/36 - 0s - loss: 0.2555 - accuracy: 0.9073 - val_loss: 0.1932 - val_accuracy: 0.9345 - 291ms/epoch - 8ms/step
Epoch 3/20
36/36 - 0s - loss: 0.2003 - accuracy: 0.9290 - val_loss: 0.1630 - val_accuracy: 0.9436 - 279ms/epoch - 8ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1743 - accuracy: 0.9387 - val_loss: 0.1594 - val_accuracy: 0.9431 - 289ms/epoch - 8ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1601 - accuracy: 0.9434 - val_loss: 0.1554 - val_accuracy: 0.9431 - 326ms/epoch - 9ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1509 - accuracy: 0.9446 - val_loss: 0.1310 - val_accuracy: 0.9502 - 284ms/epoch - 8ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1453 - accuracy: 0.9474 - val_loss: 0.1343 - val_accuracy: 0.9502 - 407ms/epoch - 11ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1410 - accuracy: 0.9485 - val_loss: 0.1286 - val_accuracy: 0.9497 - 446ms/epoch - 12ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1351 - accuracy: 0.9522 - val_loss: 0.1337 - val_accuracy: 0.9497 - 449ms/epoch - 12ms/step
Epoch 10/20
36/36 - 1s - loss: 0.1325 - accuracy: 0.9522 - val_loss: 0.1243 - val_accuracy: 0.9533 - 507ms/epoch - 14ms/step
Epoch 11/20
36/36 - 0s - loss: 0.1243 - accuracy: 0.9544 - val_loss: 0.1314 - val_accuracy: 0.9502 - 456ms/epoch - 13ms/step
Epoch 12/20
36/36 - 0s - loss: 0.1245 - accuracy: 0.9533 - val_loss: 0.1282 - val_accuracy: 0.9517 - 463ms/epoch - 13ms/step
Epoch 13/20
36/36 - 0s - loss: 0.1215 - accuracy: 0.9546 - val_loss: 0.1215 - val_accuracy: 0.9543 - 449ms/epoch - 12ms/step
Epoch 14/20
36/36 - 0s - loss: 0.1178 - accuracy: 0.9580 - val_loss: 0.1188 - val_accuracy: 0.9573 - 471ms/epoch - 13ms/step
Epoch 15/20
36/36 - 0s - loss: 0.1160 - accuracy: 0.9555 - val_loss: 0.1236 - val_accuracy: 0.9548 - 485ms/epoch - 13ms/step
Epoch 16/20
36/36 - 0s - loss: 0.1102 - accuracy: 0.9572 - val_loss: 0.1197 - val_accuracy: 0.9548 - 471ms/epoch - 13ms/step
Epoch 17/20
36/36 - 0s - loss: 0.1086 - accuracy: 0.9574 - val_loss: 0.1222 - val_accuracy: 0.9522 - 389ms/epoch - 11ms/step
Epoch 18/20
36/36 - 0s - loss: 0.1082 - accuracy: 0.9580 - val_loss: 0.1034 - val_accuracy: 0.9578 - 280ms/epoch - 8ms/step
Epoch 19/20
36/36 - 0s - loss: 0.1093 - accuracy: 0.9573 - val_loss: 0.1137 - val_accuracy: 0.9553 - 272ms/epoch - 8ms/step
Epoch 20/20
36/36 - 0s - loss: 0.1061 - accuracy: 0.9585 - val_loss: 0.0990 - val_accuracy: 0.9619 - 286ms/epoch - 8ms/step
Validation accuracy for Model of [Learning Rate 0.001 | Num Filters 32 | Dropout Rate 0.7 | FilterSize 3]: 0.9618902206420898
Epoch 1/20
36/36 - 1s - loss: 0.4929 - accuracy: 0.7904 - val_loss: 0.3399 - val_accuracy: 0.8664 - 1s/epoch - 34ms/step
Epoch 2/20
36/36 - 0s - loss: 0.2934 - accuracy: 0.8975 - val_loss: 0.2357 - val_accuracy: 0.9268 - 309ms/epoch - 9ms/step
Epoch 3/20
36/36 - 0s - loss: 0.2173 - accuracy: 0.9273 - val_loss: 0.1936 - val_accuracy: 0.9314 - 281ms/epoch - 8ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1882 - accuracy: 0.9342 - val_loss: 0.1623 - val_accuracy: 0.9441 - 281ms/epoch - 8ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1633 - accuracy: 0.9422 - val_loss: 0.1493 - val_accuracy: 0.9441 - 284ms/epoch - 8ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1490 - accuracy: 0.9442 - val_loss: 0.1381 - val_accuracy: 0.9507 - 290ms/epoch - 8ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1388 - accuracy: 0.9486 - val_loss: 0.1235 - val_accuracy: 0.9533 - 275ms/epoch - 8ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1324 - accuracy: 0.9507 - val_loss: 0.1489 - val_accuracy: 0.9446 - 298ms/epoch - 8ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1304 - accuracy: 0.9507 - val_loss: 0.1307 - val_accuracy: 0.9517 - 269ms/epoch - 7ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1205 - accuracy: 0.9539 - val_loss: 0.1247 - val_accuracy: 0.9517 - 277ms/epoch - 8ms/step
Epoch 11/20
36/36 - 0s - loss: 0.1161 - accuracy: 0.9576 - val_loss: 0.1193 - val_accuracy: 0.9548 - 337ms/epoch - 9ms/step
Epoch 12/20
36/36 - 0s - loss: 0.1146 - accuracy: 0.9560 - val_loss: 0.1177 - val_accuracy: 0.9543 - 269ms/epoch - 7ms/step
Epoch 13/20
36/36 - 0s - loss: 0.1096 - accuracy: 0.9563 - val_loss: 0.1132 - val_accuracy: 0.9578 - 272ms/epoch - 8ms/step
Epoch 14/20
36/36 - 0s - loss: 0.1064 - accuracy: 0.9578 - val_loss: 0.1068 - val_accuracy: 0.9599 - 268ms/epoch - 7ms/step
Epoch 15/20
36/36 - 0s - loss: 0.1062 - accuracy: 0.9569 - val_loss: 0.0993 - val_accuracy: 0.9599 - 287ms/epoch - 8ms/step
Epoch 16/20
36/36 - 0s - loss: 0.1051 - accuracy: 0.9589 - val_loss: 0.1132 - val_accuracy: 0.9543 - 269ms/epoch - 7ms/step
Epoch 17/20
36/36 - 0s - loss: 0.1004 - accuracy: 0.9614 - val_loss: 0.1070 - val_accuracy: 0.9593 - 273ms/epoch - 8ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0996 - accuracy: 0.9596 - val_loss: 0.1012 - val_accuracy: 0.9614 - 278ms/epoch - 8ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0966 - accuracy: 0.9603 - val_loss: 0.1168 - val_accuracy: 0.9507 - 292ms/epoch - 8ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0938 - accuracy: 0.9631 - val_loss: 0.0994 - val_accuracy: 0.9624 - 282ms/epoch - 8ms/step
Validation accuracy for Model of [Learning Rate 0.001 | Num Filters 32 | Dropout Rate 0.7 | FilterSize 5]: 0.9623983502388
Epoch 1/20
36/36 - 1s - loss: 0.5195 - accuracy: 0.7626 - val_loss: 0.3782 - val_accuracy: 0.8521 - 1s/epoch - 41ms/step
Epoch 2/20
36/36 - 0s - loss: 0.3182 - accuracy: 0.8860 - val_loss: 0.2327 - val_accuracy: 0.9212 - 275ms/epoch - 8ms/step
Epoch 3/20
36/36 - 0s - loss: 0.2274 - accuracy: 0.9194 - val_loss: 0.1837 - val_accuracy: 0.9360 - 267ms/epoch - 7ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1858 - accuracy: 0.9327 - val_loss: 0.1706 - val_accuracy: 0.9400 - 322ms/epoch - 9ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1628 - accuracy: 0.9408 - val_loss: 0.1467 - val_accuracy: 0.9451 - 268ms/epoch - 7ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1522 - accuracy: 0.9442 - val_loss: 0.1313 - val_accuracy: 0.9527 - 268ms/epoch - 7ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1409 - accuracy: 0.9498 - val_loss: 0.1295 - val_accuracy: 0.9548 - 285ms/epoch - 8ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1287 - accuracy: 0.9529 - val_loss: 0.1358 - val_accuracy: 0.9482 - 266ms/epoch - 7ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1290 - accuracy: 0.9536 - val_loss: 0.1249 - val_accuracy: 0.9543 - 275ms/epoch - 8ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1239 - accuracy: 0.9556 - val_loss: 0.1163 - val_accuracy: 0.9583 - 266ms/epoch - 7ms/step
Epoch 11/20
36/36 - 0s - loss: 0.1177 - accuracy: 0.9545 - val_loss: 0.1114 - val_accuracy: 0.9604 - 287ms/epoch - 8ms/step
Epoch 12/20
36/36 - 0s - loss: 0.1164 - accuracy: 0.9571 - val_loss: 0.1277 - val_accuracy: 0.9517 - 271ms/epoch - 8ms/step
Epoch 13/20
36/36 - 0s - loss: 0.1091 - accuracy: 0.9590 - val_loss: 0.1134 - val_accuracy: 0.9573 - 269ms/epoch - 7ms/step
Epoch 14/20
36/36 - 0s - loss: 0.1081 - accuracy: 0.9600 - val_loss: 0.1195 - val_accuracy: 0.9538 - 276ms/epoch - 8ms/step
Epoch 15/20
36/36 - 0s - loss: 0.1071 - accuracy: 0.9580 - val_loss: 0.1088 - val_accuracy: 0.9553 - 289ms/epoch - 8ms/step
Epoch 16/20
36/36 - 0s - loss: 0.1033 - accuracy: 0.9606 - val_loss: 0.1161 - val_accuracy: 0.9548 - 284ms/epoch - 8ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0996 - accuracy: 0.9615 - val_loss: 0.1335 - val_accuracy: 0.9456 - 267ms/epoch - 7ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0983 - accuracy: 0.9613 - val_loss: 0.1031 - val_accuracy: 0.9604 - 277ms/epoch - 8ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0966 - accuracy: 0.9619 - val_loss: 0.1249 - val_accuracy: 0.9512 - 271ms/epoch - 8ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0975 - accuracy: 0.9632 - val_loss: 0.1037 - val_accuracy: 0.9593 - 270ms/epoch - 7ms/step
Validation accuracy for Model of [Learning Rate 0.001 | Num Filters 32 | Dropout Rate 0.7 | FilterSize 7]: 0.9593495726585388
Epoch 1/20
36/36 - 1s - loss: 0.3899 - accuracy: 0.8417 - val_loss: 0.2616 - val_accuracy: 0.9075 - 1s/epoch - 37ms/step
Epoch 2/20
36/36 - 0s - loss: 0.2042 - accuracy: 0.9291 - val_loss: 0.2014 - val_accuracy: 0.9284 - 387ms/epoch - 11ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1625 - accuracy: 0.9426 - val_loss: 0.1428 - val_accuracy: 0.9497 - 424ms/epoch - 12ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1466 - accuracy: 0.9477 - val_loss: 0.1440 - val_accuracy: 0.9431 - 382ms/epoch - 11ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1342 - accuracy: 0.9502 - val_loss: 0.1464 - val_accuracy: 0.9446 - 410ms/epoch - 11ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1267 - accuracy: 0.9539 - val_loss: 0.1504 - val_accuracy: 0.9431 - 389ms/epoch - 11ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1209 - accuracy: 0.9552 - val_loss: 0.1174 - val_accuracy: 0.9548 - 394ms/epoch - 11ms/step
Epoch 8/20
36/36 - 1s - loss: 0.1180 - accuracy: 0.9551 - val_loss: 0.1055 - val_accuracy: 0.9578 - 511ms/epoch - 14ms/step
Epoch 9/20
36/36 - 1s - loss: 0.1127 - accuracy: 0.9561 - val_loss: 0.1130 - val_accuracy: 0.9538 - 635ms/epoch - 18ms/step
Epoch 10/20
36/36 - 1s - loss: 0.1091 - accuracy: 0.9571 - val_loss: 0.1062 - val_accuracy: 0.9563 - 655ms/epoch - 18ms/step
Epoch 11/20
36/36 - 1s - loss: 0.1009 - accuracy: 0.9590 - val_loss: 0.1248 - val_accuracy: 0.9477 - 646ms/epoch - 18ms/step
Epoch 12/20
36/36 - 1s - loss: 0.0980 - accuracy: 0.9619 - val_loss: 0.1139 - val_accuracy: 0.9548 - 647ms/epoch - 18ms/step
Epoch 13/20
36/36 - 1s - loss: 0.0958 - accuracy: 0.9610 - val_loss: 0.1156 - val_accuracy: 0.9522 - 622ms/epoch - 17ms/step
Epoch 14/20
36/36 - 1s - loss: 0.0919 - accuracy: 0.9635 - val_loss: 0.1000 - val_accuracy: 0.9639 - 614ms/epoch - 17ms/step
Epoch 15/20
36/36 - 1s - loss: 0.0866 - accuracy: 0.9650 - val_loss: 0.1085 - val_accuracy: 0.9568 - 643ms/epoch - 18ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0847 - accuracy: 0.9658 - val_loss: 0.1059 - val_accuracy: 0.9583 - 464ms/epoch - 13ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0843 - accuracy: 0.9660 - val_loss: 0.1282 - val_accuracy: 0.9461 - 382ms/epoch - 11ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0847 - accuracy: 0.9663 - val_loss: 0.0984 - val_accuracy: 0.9624 - 409ms/epoch - 11ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0823 - accuracy: 0.9676 - val_loss: 0.0892 - val_accuracy: 0.9644 - 374ms/epoch - 10ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0774 - accuracy: 0.9683 - val_loss: 0.0943 - val_accuracy: 0.9639 - 421ms/epoch - 12ms/step
Validation accuracy for Model of [Learning Rate 0.001 | Num Filters 64 | Dropout Rate 0.3 | FilterSize 3]: 0.9639227390289307
Epoch 1/20
36/36 - 1s - loss: 0.4230 - accuracy: 0.8305 - val_loss: 0.2711 - val_accuracy: 0.9136 - 1s/epoch - 37ms/step
Epoch 2/20
36/36 - 0s - loss: 0.2237 - accuracy: 0.9213 - val_loss: 0.1889 - val_accuracy: 0.9304 - 419ms/epoch - 12ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1688 - accuracy: 0.9401 - val_loss: 0.1492 - val_accuracy: 0.9446 - 369ms/epoch - 10ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1477 - accuracy: 0.9453 - val_loss: 0.1314 - val_accuracy: 0.9502 - 415ms/epoch - 12ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1353 - accuracy: 0.9498 - val_loss: 0.1260 - val_accuracy: 0.9548 - 386ms/epoch - 11ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1212 - accuracy: 0.9545 - val_loss: 0.1251 - val_accuracy: 0.9522 - 415ms/epoch - 12ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1151 - accuracy: 0.9566 - val_loss: 0.1177 - val_accuracy: 0.9543 - 430ms/epoch - 12ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1082 - accuracy: 0.9589 - val_loss: 0.1071 - val_accuracy: 0.9588 - 378ms/epoch - 10ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1050 - accuracy: 0.9600 - val_loss: 0.1273 - val_accuracy: 0.9517 - 382ms/epoch - 11ms/step
Epoch 10/20
36/36 - 0s - loss: 0.0984 - accuracy: 0.9607 - val_loss: 0.1066 - val_accuracy: 0.9558 - 429ms/epoch - 12ms/step
Epoch 11/20
36/36 - 0s - loss: 0.0973 - accuracy: 0.9617 - val_loss: 0.1073 - val_accuracy: 0.9578 - 410ms/epoch - 11ms/step
Epoch 12/20
36/36 - 0s - loss: 0.0911 - accuracy: 0.9642 - val_loss: 0.0938 - val_accuracy: 0.9649 - 386ms/epoch - 11ms/step
Epoch 13/20
36/36 - 0s - loss: 0.0910 - accuracy: 0.9636 - val_loss: 0.1054 - val_accuracy: 0.9568 - 362ms/epoch - 10ms/step
Epoch 14/20
36/36 - 0s - loss: 0.0866 - accuracy: 0.9658 - val_loss: 0.1161 - val_accuracy: 0.9502 - 362ms/epoch - 10ms/step
Epoch 15/20
36/36 - 0s - loss: 0.0825 - accuracy: 0.9659 - val_loss: 0.1014 - val_accuracy: 0.9593 - 423ms/epoch - 12ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0809 - accuracy: 0.9668 - val_loss: 0.1006 - val_accuracy: 0.9599 - 381ms/epoch - 11ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0793 - accuracy: 0.9689 - val_loss: 0.1000 - val_accuracy: 0.9588 - 369ms/epoch - 10ms/step
Epoch 18/20
36/36 - 1s - loss: 0.0771 - accuracy: 0.9701 - val_loss: 0.0969 - val_accuracy: 0.9624 - 586ms/epoch - 16ms/step
Epoch 19/20
36/36 - 1s - loss: 0.0747 - accuracy: 0.9703 - val_loss: 0.0899 - val_accuracy: 0.9675 - 588ms/epoch - 16ms/step
Epoch 20/20
36/36 - 1s - loss: 0.0768 - accuracy: 0.9689 - val_loss: 0.0937 - val_accuracy: 0.9619 - 557ms/epoch - 15ms/step
Validation accuracy for Model of [Learning Rate 0.001 | Num Filters 64 | Dropout Rate 0.3 | FilterSize 5]: 0.9618902206420898
Epoch 1/20
36/36 - 2s - loss: 0.4356 - accuracy: 0.8142 - val_loss: 0.2544 - val_accuracy: 0.9131 - 2s/epoch - 55ms/step
Epoch 2/20
36/36 - 1s - loss: 0.2231 - accuracy: 0.9197 - val_loss: 0.1818 - val_accuracy: 0.9319 - 587ms/epoch - 16ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1654 - accuracy: 0.9406 - val_loss: 0.1706 - val_accuracy: 0.9339 - 528ms/epoch - 15ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1423 - accuracy: 0.9483 - val_loss: 0.1242 - val_accuracy: 0.9573 - 366ms/epoch - 10ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1331 - accuracy: 0.9513 - val_loss: 0.1414 - val_accuracy: 0.9466 - 353ms/epoch - 10ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1184 - accuracy: 0.9538 - val_loss: 0.1140 - val_accuracy: 0.9568 - 352ms/epoch - 10ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1123 - accuracy: 0.9549 - val_loss: 0.1322 - val_accuracy: 0.9487 - 403ms/epoch - 11ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1053 - accuracy: 0.9592 - val_loss: 0.0997 - val_accuracy: 0.9614 - 387ms/epoch - 11ms/step
Epoch 9/20
36/36 - 0s - loss: 0.0985 - accuracy: 0.9607 - val_loss: 0.1052 - val_accuracy: 0.9583 - 395ms/epoch - 11ms/step
Epoch 10/20
36/36 - 0s - loss: 0.0959 - accuracy: 0.9607 - val_loss: 0.1264 - val_accuracy: 0.9466 - 359ms/epoch - 10ms/step
Epoch 11/20
36/36 - 0s - loss: 0.0915 - accuracy: 0.9629 - val_loss: 0.1112 - val_accuracy: 0.9527 - 359ms/epoch - 10ms/step
Epoch 12/20
36/36 - 0s - loss: 0.0896 - accuracy: 0.9644 - val_loss: 0.1024 - val_accuracy: 0.9573 - 401ms/epoch - 11ms/step
Epoch 13/20
36/36 - 0s - loss: 0.0832 - accuracy: 0.9670 - val_loss: 0.0986 - val_accuracy: 0.9604 - 354ms/epoch - 10ms/step
Epoch 14/20
36/36 - 0s - loss: 0.0800 - accuracy: 0.9684 - val_loss: 0.0970 - val_accuracy: 0.9599 - 386ms/epoch - 11ms/step
Epoch 15/20
36/36 - 0s - loss: 0.0791 - accuracy: 0.9678 - val_loss: 0.0973 - val_accuracy: 0.9624 - 361ms/epoch - 10ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0768 - accuracy: 0.9689 - val_loss: 0.0920 - val_accuracy: 0.9649 - 356ms/epoch - 10ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0727 - accuracy: 0.9710 - val_loss: 0.0943 - val_accuracy: 0.9629 - 399ms/epoch - 11ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0734 - accuracy: 0.9703 - val_loss: 0.1101 - val_accuracy: 0.9522 - 378ms/epoch - 10ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0737 - accuracy: 0.9698 - val_loss: 0.0850 - val_accuracy: 0.9680 - 365ms/epoch - 10ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0684 - accuracy: 0.9718 - val_loss: 0.0978 - val_accuracy: 0.9604 - 365ms/epoch - 10ms/step
Validation accuracy for Model of [Learning Rate 0.001 | Num Filters 64 | Dropout Rate 0.3 | FilterSize 7]: 0.9603658318519592
Epoch 1/20
36/36 - 1s - loss: 0.4035 - accuracy: 0.8390 - val_loss: 0.2402 - val_accuracy: 0.9151 - 1s/epoch - 38ms/step
Epoch 2/20
36/36 - 0s - loss: 0.2206 - accuracy: 0.9230 - val_loss: 0.1940 - val_accuracy: 0.9355 - 430ms/epoch - 12ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1740 - accuracy: 0.9410 - val_loss: 0.1379 - val_accuracy: 0.9507 - 433ms/epoch - 12ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1541 - accuracy: 0.9454 - val_loss: 0.1395 - val_accuracy: 0.9472 - 435ms/epoch - 12ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1446 - accuracy: 0.9478 - val_loss: 0.1381 - val_accuracy: 0.9456 - 394ms/epoch - 11ms/step
Epoch 6/20
36/36 - 1s - loss: 0.1317 - accuracy: 0.9528 - val_loss: 0.1174 - val_accuracy: 0.9553 - 510ms/epoch - 14ms/step
Epoch 7/20
36/36 - 1s - loss: 0.1271 - accuracy: 0.9524 - val_loss: 0.1234 - val_accuracy: 0.9533 - 626ms/epoch - 17ms/step
Epoch 8/20
36/36 - 1s - loss: 0.1213 - accuracy: 0.9541 - val_loss: 0.1129 - val_accuracy: 0.9573 - 675ms/epoch - 19ms/step
Epoch 9/20
36/36 - 1s - loss: 0.1155 - accuracy: 0.9547 - val_loss: 0.1171 - val_accuracy: 0.9578 - 845ms/epoch - 23ms/step
Epoch 10/20
36/36 - 1s - loss: 0.1128 - accuracy: 0.9566 - val_loss: 0.1323 - val_accuracy: 0.9472 - 775ms/epoch - 22ms/step
Epoch 11/20
36/36 - 1s - loss: 0.1087 - accuracy: 0.9570 - val_loss: 0.1089 - val_accuracy: 0.9553 - 771ms/epoch - 21ms/step
Epoch 12/20
36/36 - 1s - loss: 0.1037 - accuracy: 0.9584 - val_loss: 0.1080 - val_accuracy: 0.9573 - 764ms/epoch - 21ms/step
Epoch 13/20
36/36 - 1s - loss: 0.1029 - accuracy: 0.9591 - val_loss: 0.1113 - val_accuracy: 0.9543 - 756ms/epoch - 21ms/step
Epoch 14/20
36/36 - 1s - loss: 0.0995 - accuracy: 0.9601 - val_loss: 0.1024 - val_accuracy: 0.9604 - 743ms/epoch - 21ms/step
Epoch 15/20
36/36 - 1s - loss: 0.0953 - accuracy: 0.9621 - val_loss: 0.1135 - val_accuracy: 0.9553 - 766ms/epoch - 21ms/step
Epoch 16/20
36/36 - 1s - loss: 0.0949 - accuracy: 0.9615 - val_loss: 0.0970 - val_accuracy: 0.9624 - 723ms/epoch - 20ms/step
Epoch 17/20
36/36 - 1s - loss: 0.0908 - accuracy: 0.9626 - val_loss: 0.1041 - val_accuracy: 0.9588 - 715ms/epoch - 20ms/step
Epoch 18/20
36/36 - 1s - loss: 0.0891 - accuracy: 0.9634 - val_loss: 0.0965 - val_accuracy: 0.9629 - 612ms/epoch - 17ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0866 - accuracy: 0.9649 - val_loss: 0.0986 - val_accuracy: 0.9599 - 403ms/epoch - 11ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0868 - accuracy: 0.9640 - val_loss: 0.0926 - val_accuracy: 0.9654 - 406ms/epoch - 11ms/step
Validation accuracy for Model of [Learning Rate 0.001 | Num Filters 64 | Dropout Rate 0.5 | FilterSize 3]: 0.9654471278190613
Epoch 1/20
36/36 - 1s - loss: 0.4323 - accuracy: 0.8211 - val_loss: 0.2589 - val_accuracy: 0.9162 - 1s/epoch - 37ms/step
Epoch 2/20
36/36 - 0s - loss: 0.2335 - accuracy: 0.9160 - val_loss: 0.1955 - val_accuracy: 0.9304 - 401ms/epoch - 11ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1761 - accuracy: 0.9370 - val_loss: 0.1518 - val_accuracy: 0.9451 - 370ms/epoch - 10ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1498 - accuracy: 0.9453 - val_loss: 0.1239 - val_accuracy: 0.9578 - 381ms/epoch - 11ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1398 - accuracy: 0.9497 - val_loss: 0.1282 - val_accuracy: 0.9548 - 363ms/epoch - 10ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1268 - accuracy: 0.9548 - val_loss: 0.1472 - val_accuracy: 0.9451 - 386ms/epoch - 11ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1231 - accuracy: 0.9552 - val_loss: 0.1160 - val_accuracy: 0.9573 - 412ms/epoch - 11ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1133 - accuracy: 0.9568 - val_loss: 0.1656 - val_accuracy: 0.9360 - 371ms/epoch - 10ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1079 - accuracy: 0.9568 - val_loss: 0.1038 - val_accuracy: 0.9604 - 424ms/epoch - 12ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1040 - accuracy: 0.9599 - val_loss: 0.1249 - val_accuracy: 0.9502 - 407ms/epoch - 11ms/step
Epoch 11/20
36/36 - 0s - loss: 0.0999 - accuracy: 0.9608 - val_loss: 0.1253 - val_accuracy: 0.9482 - 373ms/epoch - 10ms/step
Epoch 12/20
36/36 - 0s - loss: 0.0977 - accuracy: 0.9607 - val_loss: 0.0963 - val_accuracy: 0.9614 - 418ms/epoch - 12ms/step
Epoch 13/20
36/36 - 0s - loss: 0.0935 - accuracy: 0.9632 - val_loss: 0.1080 - val_accuracy: 0.9563 - 357ms/epoch - 10ms/step
Epoch 14/20
36/36 - 0s - loss: 0.0919 - accuracy: 0.9616 - val_loss: 0.1169 - val_accuracy: 0.9482 - 370ms/epoch - 10ms/step
Epoch 15/20
36/36 - 0s - loss: 0.0877 - accuracy: 0.9657 - val_loss: 0.1028 - val_accuracy: 0.9604 - 401ms/epoch - 11ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0856 - accuracy: 0.9659 - val_loss: 0.1194 - val_accuracy: 0.9507 - 378ms/epoch - 10ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0868 - accuracy: 0.9652 - val_loss: 0.0992 - val_accuracy: 0.9609 - 425ms/epoch - 12ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0835 - accuracy: 0.9670 - val_loss: 0.0910 - val_accuracy: 0.9649 - 361ms/epoch - 10ms/step
Epoch 19/20
36/36 - 1s - loss: 0.0815 - accuracy: 0.9681 - val_loss: 0.0874 - val_accuracy: 0.9660 - 587ms/epoch - 16ms/step
Epoch 20/20
36/36 - 1s - loss: 0.0764 - accuracy: 0.9692 - val_loss: 0.0924 - val_accuracy: 0.9629 - 602ms/epoch - 17ms/step
Validation accuracy for Model of [Learning Rate 0.001 | Num Filters 64 | Dropout Rate 0.5 | FilterSize 5]: 0.9629064798355103
Epoch 1/20
36/36 - 2s - loss: 0.4482 - accuracy: 0.8148 - val_loss: 0.3166 - val_accuracy: 0.8674 - 2s/epoch - 56ms/step
Epoch 2/20
36/36 - 1s - loss: 0.2495 - accuracy: 0.9110 - val_loss: 0.1915 - val_accuracy: 0.9375 - 590ms/epoch - 16ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1800 - accuracy: 0.9382 - val_loss: 0.1526 - val_accuracy: 0.9441 - 589ms/epoch - 16ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1516 - accuracy: 0.9450 - val_loss: 0.1416 - val_accuracy: 0.9527 - 423ms/epoch - 12ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1322 - accuracy: 0.9521 - val_loss: 0.1267 - val_accuracy: 0.9558 - 388ms/epoch - 11ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1246 - accuracy: 0.9553 - val_loss: 0.1292 - val_accuracy: 0.9533 - 397ms/epoch - 11ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1147 - accuracy: 0.9572 - val_loss: 0.1080 - val_accuracy: 0.9609 - 353ms/epoch - 10ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1078 - accuracy: 0.9588 - val_loss: 0.1052 - val_accuracy: 0.9609 - 393ms/epoch - 11ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1023 - accuracy: 0.9604 - val_loss: 0.1124 - val_accuracy: 0.9563 - 399ms/epoch - 11ms/step
Epoch 10/20
36/36 - 0s - loss: 0.0983 - accuracy: 0.9611 - val_loss: 0.1256 - val_accuracy: 0.9517 - 360ms/epoch - 10ms/step
Epoch 11/20
36/36 - 0s - loss: 0.0972 - accuracy: 0.9614 - val_loss: 0.1023 - val_accuracy: 0.9604 - 349ms/epoch - 10ms/step
Epoch 12/20
36/36 - 0s - loss: 0.0918 - accuracy: 0.9616 - val_loss: 0.1028 - val_accuracy: 0.9588 - 359ms/epoch - 10ms/step
Epoch 13/20
36/36 - 0s - loss: 0.0889 - accuracy: 0.9632 - val_loss: 0.1039 - val_accuracy: 0.9563 - 393ms/epoch - 11ms/step
Epoch 14/20
36/36 - 0s - loss: 0.0857 - accuracy: 0.9666 - val_loss: 0.1052 - val_accuracy: 0.9573 - 382ms/epoch - 11ms/step
Epoch 15/20
36/36 - 0s - loss: 0.0837 - accuracy: 0.9659 - val_loss: 0.1078 - val_accuracy: 0.9553 - 351ms/epoch - 10ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0835 - accuracy: 0.9667 - val_loss: 0.1044 - val_accuracy: 0.9558 - 396ms/epoch - 11ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0773 - accuracy: 0.9690 - val_loss: 0.0982 - val_accuracy: 0.9639 - 370ms/epoch - 10ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0767 - accuracy: 0.9695 - val_loss: 0.1034 - val_accuracy: 0.9583 - 354ms/epoch - 10ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0753 - accuracy: 0.9683 - val_loss: 0.0990 - val_accuracy: 0.9614 - 352ms/epoch - 10ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0764 - accuracy: 0.9691 - val_loss: 0.0973 - val_accuracy: 0.9593 - 377ms/epoch - 10ms/step
Validation accuracy for Model of [Learning Rate 0.001 | Num Filters 64 | Dropout Rate 0.5 | FilterSize 7]: 0.9593495726585388
Epoch 1/20
36/36 - 3s - loss: 0.4090 - accuracy: 0.8370 - val_loss: 0.2542 - val_accuracy: 0.9172 - 3s/epoch - 70ms/step
Epoch 2/20
36/36 - 1s - loss: 0.2255 - accuracy: 0.9196 - val_loss: 0.1696 - val_accuracy: 0.9436 - 641ms/epoch - 18ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1764 - accuracy: 0.9388 - val_loss: 0.1561 - val_accuracy: 0.9461 - 643ms/epoch - 18ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1601 - accuracy: 0.9428 - val_loss: 0.1326 - val_accuracy: 0.9527 - 623ms/epoch - 17ms/step
Epoch 5/20
36/36 - 1s - loss: 0.1508 - accuracy: 0.9456 - val_loss: 0.1364 - val_accuracy: 0.9492 - 634ms/epoch - 18ms/step
Epoch 6/20
36/36 - 1s - loss: 0.1420 - accuracy: 0.9479 - val_loss: 0.1287 - val_accuracy: 0.9507 - 620ms/epoch - 17ms/step
Epoch 7/20
36/36 - 1s - loss: 0.1359 - accuracy: 0.9480 - val_loss: 0.1214 - val_accuracy: 0.9543 - 594ms/epoch - 16ms/step
Epoch 8/20
36/36 - 1s - loss: 0.1290 - accuracy: 0.9534 - val_loss: 0.1222 - val_accuracy: 0.9512 - 554ms/epoch - 15ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1239 - accuracy: 0.9531 - val_loss: 0.1365 - val_accuracy: 0.9477 - 439ms/epoch - 12ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1202 - accuracy: 0.9520 - val_loss: 0.1148 - val_accuracy: 0.9538 - 422ms/epoch - 12ms/step
Epoch 11/20
36/36 - 0s - loss: 0.1143 - accuracy: 0.9549 - val_loss: 0.1115 - val_accuracy: 0.9558 - 427ms/epoch - 12ms/step
Epoch 12/20
36/36 - 0s - loss: 0.1166 - accuracy: 0.9539 - val_loss: 0.1168 - val_accuracy: 0.9533 - 442ms/epoch - 12ms/step
Epoch 13/20
36/36 - 0s - loss: 0.1099 - accuracy: 0.9562 - val_loss: 0.1206 - val_accuracy: 0.9512 - 423ms/epoch - 12ms/step
Epoch 14/20
36/36 - 0s - loss: 0.1084 - accuracy: 0.9581 - val_loss: 0.1023 - val_accuracy: 0.9604 - 445ms/epoch - 12ms/step
Epoch 15/20
36/36 - 0s - loss: 0.1030 - accuracy: 0.9592 - val_loss: 0.1073 - val_accuracy: 0.9568 - 430ms/epoch - 12ms/step
Epoch 16/20
36/36 - 0s - loss: 0.1026 - accuracy: 0.9594 - val_loss: 0.1204 - val_accuracy: 0.9497 - 398ms/epoch - 11ms/step
Epoch 17/20
36/36 - 0s - loss: 0.1028 - accuracy: 0.9593 - val_loss: 0.1069 - val_accuracy: 0.9543 - 423ms/epoch - 12ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0978 - accuracy: 0.9600 - val_loss: 0.1106 - val_accuracy: 0.9563 - 401ms/epoch - 11ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0961 - accuracy: 0.9600 - val_loss: 0.1206 - val_accuracy: 0.9497 - 409ms/epoch - 11ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0974 - accuracy: 0.9606 - val_loss: 0.1013 - val_accuracy: 0.9588 - 392ms/epoch - 11ms/step
Validation accuracy for Model of [Learning Rate 0.001 | Num Filters 64 | Dropout Rate 0.7 | FilterSize 3]: 0.9588414430618286
Epoch 1/20
36/36 - 1s - loss: 0.4456 - accuracy: 0.8052 - val_loss: 0.2915 - val_accuracy: 0.8953 - 1s/epoch - 38ms/step
Epoch 2/20
36/36 - 0s - loss: 0.2564 - accuracy: 0.9090 - val_loss: 0.1868 - val_accuracy: 0.9355 - 415ms/epoch - 12ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1931 - accuracy: 0.9313 - val_loss: 0.1617 - val_accuracy: 0.9426 - 413ms/epoch - 11ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1640 - accuracy: 0.9415 - val_loss: 0.1607 - val_accuracy: 0.9390 - 385ms/epoch - 11ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1483 - accuracy: 0.9456 - val_loss: 0.1230 - val_accuracy: 0.9553 - 406ms/epoch - 11ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1365 - accuracy: 0.9498 - val_loss: 0.1217 - val_accuracy: 0.9553 - 385ms/epoch - 11ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1299 - accuracy: 0.9516 - val_loss: 0.1400 - val_accuracy: 0.9487 - 363ms/epoch - 10ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1224 - accuracy: 0.9524 - val_loss: 0.1321 - val_accuracy: 0.9512 - 380ms/epoch - 11ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1159 - accuracy: 0.9549 - val_loss: 0.1131 - val_accuracy: 0.9563 - 431ms/epoch - 12ms/step
Epoch 10/20
36/36 - 1s - loss: 0.1146 - accuracy: 0.9559 - val_loss: 0.1235 - val_accuracy: 0.9522 - 597ms/epoch - 17ms/step
Epoch 11/20
36/36 - 1s - loss: 0.1119 - accuracy: 0.9563 - val_loss: 0.1089 - val_accuracy: 0.9609 - 593ms/epoch - 16ms/step
Epoch 12/20
36/36 - 1s - loss: 0.1083 - accuracy: 0.9603 - val_loss: 0.1110 - val_accuracy: 0.9548 - 627ms/epoch - 17ms/step
Epoch 13/20
36/36 - 1s - loss: 0.1040 - accuracy: 0.9593 - val_loss: 0.1071 - val_accuracy: 0.9558 - 586ms/epoch - 16ms/step
Epoch 14/20
36/36 - 1s - loss: 0.1004 - accuracy: 0.9604 - val_loss: 0.1061 - val_accuracy: 0.9578 - 602ms/epoch - 17ms/step
Epoch 15/20
36/36 - 1s - loss: 0.0962 - accuracy: 0.9611 - val_loss: 0.1162 - val_accuracy: 0.9502 - 604ms/epoch - 17ms/step
Epoch 16/20
36/36 - 1s - loss: 0.0962 - accuracy: 0.9621 - val_loss: 0.1062 - val_accuracy: 0.9578 - 598ms/epoch - 17ms/step
Epoch 17/20
36/36 - 1s - loss: 0.0954 - accuracy: 0.9603 - val_loss: 0.1044 - val_accuracy: 0.9583 - 621ms/epoch - 17ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0918 - accuracy: 0.9621 - val_loss: 0.0921 - val_accuracy: 0.9665 - 440ms/epoch - 12ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0944 - accuracy: 0.9631 - val_loss: 0.1023 - val_accuracy: 0.9619 - 369ms/epoch - 10ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0899 - accuracy: 0.9628 - val_loss: 0.0984 - val_accuracy: 0.9624 - 374ms/epoch - 10ms/step
Validation accuracy for Model of [Learning Rate 0.001 | Num Filters 64 | Dropout Rate 0.7 | FilterSize 5]: 0.9623983502388
Epoch 1/20
36/36 - 1s - loss: 0.4698 - accuracy: 0.7940 - val_loss: 0.3291 - val_accuracy: 0.8725 - 1s/epoch - 37ms/step
Epoch 2/20
36/36 - 0s - loss: 0.2639 - accuracy: 0.9042 - val_loss: 0.1994 - val_accuracy: 0.9304 - 382ms/epoch - 11ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1927 - accuracy: 0.9317 - val_loss: 0.1735 - val_accuracy: 0.9334 - 359ms/epoch - 10ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1664 - accuracy: 0.9400 - val_loss: 0.1428 - val_accuracy: 0.9461 - 345ms/epoch - 10ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1475 - accuracy: 0.9465 - val_loss: 0.1467 - val_accuracy: 0.9436 - 352ms/epoch - 10ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1339 - accuracy: 0.9501 - val_loss: 0.1417 - val_accuracy: 0.9477 - 358ms/epoch - 10ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1262 - accuracy: 0.9544 - val_loss: 0.1369 - val_accuracy: 0.9487 - 395ms/epoch - 11ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1189 - accuracy: 0.9551 - val_loss: 0.1236 - val_accuracy: 0.9527 - 377ms/epoch - 10ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1130 - accuracy: 0.9567 - val_loss: 0.1336 - val_accuracy: 0.9466 - 397ms/epoch - 11ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1096 - accuracy: 0.9592 - val_loss: 0.1108 - val_accuracy: 0.9553 - 392ms/epoch - 11ms/step
Epoch 11/20
36/36 - 0s - loss: 0.1075 - accuracy: 0.9578 - val_loss: 0.1228 - val_accuracy: 0.9517 - 362ms/epoch - 10ms/step
Epoch 12/20
36/36 - 0s - loss: 0.1056 - accuracy: 0.9594 - val_loss: 0.1136 - val_accuracy: 0.9578 - 356ms/epoch - 10ms/step
Epoch 13/20
36/36 - 0s - loss: 0.1001 - accuracy: 0.9620 - val_loss: 0.1021 - val_accuracy: 0.9609 - 352ms/epoch - 10ms/step
Epoch 14/20
36/36 - 0s - loss: 0.0980 - accuracy: 0.9614 - val_loss: 0.1014 - val_accuracy: 0.9588 - 394ms/epoch - 11ms/step
Epoch 15/20
36/36 - 0s - loss: 0.0950 - accuracy: 0.9602 - val_loss: 0.1066 - val_accuracy: 0.9599 - 383ms/epoch - 11ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0945 - accuracy: 0.9622 - val_loss: 0.1001 - val_accuracy: 0.9588 - 387ms/epoch - 11ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0888 - accuracy: 0.9650 - val_loss: 0.1102 - val_accuracy: 0.9522 - 368ms/epoch - 10ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0907 - accuracy: 0.9632 - val_loss: 0.0996 - val_accuracy: 0.9619 - 395ms/epoch - 11ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0861 - accuracy: 0.9669 - val_loss: 0.1248 - val_accuracy: 0.9487 - 408ms/epoch - 11ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0875 - accuracy: 0.9637 - val_loss: 0.1069 - val_accuracy: 0.9558 - 347ms/epoch - 10ms/step
Validation accuracy for Model of [Learning Rate 0.001 | Num Filters 64 | Dropout Rate 0.7 | FilterSize 7]: 0.9557926654815674
Epoch 1/20
36/36 - 2s - loss: 0.3611 - accuracy: 0.8690 - val_loss: 0.1860 - val_accuracy: 0.9273 - 2s/epoch - 65ms/step
Epoch 2/20
36/36 - 1s - loss: 0.1938 - accuracy: 0.9335 - val_loss: 0.1615 - val_accuracy: 0.9395 - 994ms/epoch - 28ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1590 - accuracy: 0.9423 - val_loss: 0.1520 - val_accuracy: 0.9431 - 996ms/epoch - 28ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1450 - accuracy: 0.9466 - val_loss: 0.1378 - val_accuracy: 0.9466 - 900ms/epoch - 25ms/step
Epoch 5/20
36/36 - 1s - loss: 0.1325 - accuracy: 0.9491 - val_loss: 0.1142 - val_accuracy: 0.9563 - 604ms/epoch - 17ms/step
Epoch 6/20
36/36 - 1s - loss: 0.1212 - accuracy: 0.9526 - val_loss: 0.1342 - val_accuracy: 0.9512 - 586ms/epoch - 16ms/step
Epoch 7/20
36/36 - 1s - loss: 0.1130 - accuracy: 0.9567 - val_loss: 0.1239 - val_accuracy: 0.9522 - 597ms/epoch - 17ms/step
Epoch 8/20
36/36 - 1s - loss: 0.1104 - accuracy: 0.9573 - val_loss: 0.1226 - val_accuracy: 0.9558 - 581ms/epoch - 16ms/step
Epoch 9/20
36/36 - 1s - loss: 0.1068 - accuracy: 0.9566 - val_loss: 0.1209 - val_accuracy: 0.9533 - 612ms/epoch - 17ms/step
Epoch 10/20
36/36 - 1s - loss: 0.1015 - accuracy: 0.9597 - val_loss: 0.1131 - val_accuracy: 0.9558 - 606ms/epoch - 17ms/step
Epoch 11/20
36/36 - 1s - loss: 0.1002 - accuracy: 0.9601 - val_loss: 0.0996 - val_accuracy: 0.9614 - 588ms/epoch - 16ms/step
Epoch 12/20
36/36 - 1s - loss: 0.0965 - accuracy: 0.9604 - val_loss: 0.1063 - val_accuracy: 0.9588 - 600ms/epoch - 17ms/step
Epoch 13/20
36/36 - 1s - loss: 0.0920 - accuracy: 0.9616 - val_loss: 0.1075 - val_accuracy: 0.9583 - 617ms/epoch - 17ms/step
Epoch 14/20
36/36 - 1s - loss: 0.0873 - accuracy: 0.9639 - val_loss: 0.1011 - val_accuracy: 0.9614 - 601ms/epoch - 17ms/step
Epoch 15/20
36/36 - 1s - loss: 0.0863 - accuracy: 0.9633 - val_loss: 0.1386 - val_accuracy: 0.9431 - 605ms/epoch - 17ms/step
Epoch 16/20
36/36 - 1s - loss: 0.0867 - accuracy: 0.9651 - val_loss: 0.1125 - val_accuracy: 0.9558 - 591ms/epoch - 16ms/step
Epoch 17/20
36/36 - 1s - loss: 0.0814 - accuracy: 0.9676 - val_loss: 0.0977 - val_accuracy: 0.9624 - 610ms/epoch - 17ms/step
Epoch 18/20
36/36 - 1s - loss: 0.0787 - accuracy: 0.9672 - val_loss: 0.0970 - val_accuracy: 0.9654 - 606ms/epoch - 17ms/step
Epoch 19/20
36/36 - 1s - loss: 0.0785 - accuracy: 0.9690 - val_loss: 0.0914 - val_accuracy: 0.9685 - 600ms/epoch - 17ms/step
Epoch 20/20
36/36 - 1s - loss: 0.0756 - accuracy: 0.9701 - val_loss: 0.0836 - val_accuracy: 0.9715 - 593ms/epoch - 16ms/step
Validation accuracy for Model of [Learning Rate 0.001 | Num Filters 128 | Dropout Rate 0.3 | FilterSize 3]: 0.9715447425842285
Epoch 1/20
36/36 - 2s - loss: 0.3831 - accuracy: 0.8471 - val_loss: 0.2429 - val_accuracy: 0.9197 - 2s/epoch - 66ms/step
Epoch 2/20
36/36 - 1s - loss: 0.1939 - accuracy: 0.9340 - val_loss: 0.1755 - val_accuracy: 0.9355 - 932ms/epoch - 26ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1506 - accuracy: 0.9467 - val_loss: 0.1510 - val_accuracy: 0.9482 - 948ms/epoch - 26ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1331 - accuracy: 0.9528 - val_loss: 0.1284 - val_accuracy: 0.9543 - 757ms/epoch - 21ms/step
Epoch 5/20
36/36 - 1s - loss: 0.1170 - accuracy: 0.9549 - val_loss: 0.1221 - val_accuracy: 0.9502 - 609ms/epoch - 17ms/step
Epoch 6/20
36/36 - 1s - loss: 0.1119 - accuracy: 0.9556 - val_loss: 0.1206 - val_accuracy: 0.9548 - 594ms/epoch - 17ms/step
Epoch 7/20
36/36 - 1s - loss: 0.1062 - accuracy: 0.9574 - val_loss: 0.1259 - val_accuracy: 0.9482 - 573ms/epoch - 16ms/step
Epoch 8/20
36/36 - 1s - loss: 0.1012 - accuracy: 0.9599 - val_loss: 0.1215 - val_accuracy: 0.9507 - 611ms/epoch - 17ms/step
Epoch 9/20
36/36 - 1s - loss: 0.0928 - accuracy: 0.9639 - val_loss: 0.0941 - val_accuracy: 0.9670 - 598ms/epoch - 17ms/step
Epoch 10/20
36/36 - 1s - loss: 0.0897 - accuracy: 0.9653 - val_loss: 0.1072 - val_accuracy: 0.9568 - 583ms/epoch - 16ms/step
Epoch 11/20
36/36 - 1s - loss: 0.0846 - accuracy: 0.9663 - val_loss: 0.1067 - val_accuracy: 0.9583 - 573ms/epoch - 16ms/step
Epoch 12/20
36/36 - 1s - loss: 0.0823 - accuracy: 0.9667 - val_loss: 0.1097 - val_accuracy: 0.9553 - 586ms/epoch - 16ms/step
Epoch 13/20
36/36 - 1s - loss: 0.0846 - accuracy: 0.9660 - val_loss: 0.0903 - val_accuracy: 0.9660 - 579ms/epoch - 16ms/step
Epoch 14/20
36/36 - 1s - loss: 0.0807 - accuracy: 0.9687 - val_loss: 0.0889 - val_accuracy: 0.9685 - 602ms/epoch - 17ms/step
Epoch 15/20
36/36 - 1s - loss: 0.0755 - accuracy: 0.9697 - val_loss: 0.0948 - val_accuracy: 0.9629 - 594ms/epoch - 16ms/step
Epoch 16/20
36/36 - 1s - loss: 0.0748 - accuracy: 0.9701 - val_loss: 0.0862 - val_accuracy: 0.9700 - 577ms/epoch - 16ms/step
Epoch 17/20
36/36 - 1s - loss: 0.0722 - accuracy: 0.9706 - val_loss: 0.0844 - val_accuracy: 0.9685 - 577ms/epoch - 16ms/step
Epoch 18/20
36/36 - 1s - loss: 0.0692 - accuracy: 0.9727 - val_loss: 0.1174 - val_accuracy: 0.9477 - 676ms/epoch - 19ms/step
Epoch 19/20
36/36 - 1s - loss: 0.0685 - accuracy: 0.9734 - val_loss: 0.0944 - val_accuracy: 0.9634 - 609ms/epoch - 17ms/step
Epoch 20/20
36/36 - 1s - loss: 0.0707 - accuracy: 0.9713 - val_loss: 0.0960 - val_accuracy: 0.9639 - 595ms/epoch - 17ms/step
Validation accuracy for Model of [Learning Rate 0.001 | Num Filters 128 | Dropout Rate 0.3 | FilterSize 5]: 0.9639227390289307
Epoch 1/20
36/36 - 1s - loss: 0.3976 - accuracy: 0.8358 - val_loss: 0.2869 - val_accuracy: 0.8943 - 1s/epoch - 41ms/step
Epoch 2/20
36/36 - 1s - loss: 0.2007 - accuracy: 0.9298 - val_loss: 0.1636 - val_accuracy: 0.9400 - 533ms/epoch - 15ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1534 - accuracy: 0.9443 - val_loss: 0.1386 - val_accuracy: 0.9472 - 541ms/epoch - 15ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1343 - accuracy: 0.9511 - val_loss: 0.1687 - val_accuracy: 0.9319 - 521ms/epoch - 14ms/step
Epoch 5/20
36/36 - 1s - loss: 0.1201 - accuracy: 0.9544 - val_loss: 0.1398 - val_accuracy: 0.9466 - 518ms/epoch - 14ms/step
Epoch 6/20
36/36 - 1s - loss: 0.1115 - accuracy: 0.9571 - val_loss: 0.1385 - val_accuracy: 0.9451 - 516ms/epoch - 14ms/step
Epoch 7/20
36/36 - 1s - loss: 0.1010 - accuracy: 0.9611 - val_loss: 0.1037 - val_accuracy: 0.9614 - 518ms/epoch - 14ms/step
Epoch 8/20
36/36 - 1s - loss: 0.0956 - accuracy: 0.9618 - val_loss: 0.1133 - val_accuracy: 0.9548 - 533ms/epoch - 15ms/step
Epoch 9/20
36/36 - 1s - loss: 0.0907 - accuracy: 0.9646 - val_loss: 0.1047 - val_accuracy: 0.9604 - 508ms/epoch - 14ms/step
Epoch 10/20
36/36 - 1s - loss: 0.0884 - accuracy: 0.9647 - val_loss: 0.1347 - val_accuracy: 0.9446 - 859ms/epoch - 24ms/step
Epoch 11/20
36/36 - 1s - loss: 0.0856 - accuracy: 0.9647 - val_loss: 0.1029 - val_accuracy: 0.9593 - 869ms/epoch - 24ms/step
Epoch 12/20
36/36 - 1s - loss: 0.0819 - accuracy: 0.9668 - val_loss: 0.1030 - val_accuracy: 0.9583 - 849ms/epoch - 24ms/step
Epoch 13/20
36/36 - 1s - loss: 0.0786 - accuracy: 0.9675 - val_loss: 0.0901 - val_accuracy: 0.9670 - 852ms/epoch - 24ms/step
Epoch 14/20
36/36 - 1s - loss: 0.0782 - accuracy: 0.9678 - val_loss: 0.0895 - val_accuracy: 0.9665 - 864ms/epoch - 24ms/step
Epoch 15/20
36/36 - 1s - loss: 0.0724 - accuracy: 0.9707 - val_loss: 0.0940 - val_accuracy: 0.9629 - 724ms/epoch - 20ms/step
Epoch 16/20
36/36 - 1s - loss: 0.0729 - accuracy: 0.9698 - val_loss: 0.0887 - val_accuracy: 0.9690 - 524ms/epoch - 15ms/step
Epoch 17/20
36/36 - 1s - loss: 0.0683 - accuracy: 0.9722 - val_loss: 0.1040 - val_accuracy: 0.9553 - 515ms/epoch - 14ms/step
Epoch 18/20
36/36 - 1s - loss: 0.0668 - accuracy: 0.9735 - val_loss: 0.0963 - val_accuracy: 0.9604 - 541ms/epoch - 15ms/step
Epoch 19/20
36/36 - 1s - loss: 0.0651 - accuracy: 0.9743 - val_loss: 0.0902 - val_accuracy: 0.9660 - 528ms/epoch - 15ms/step
Epoch 20/20
36/36 - 1s - loss: 0.0640 - accuracy: 0.9742 - val_loss: 0.0941 - val_accuracy: 0.9660 - 541ms/epoch - 15ms/step
Validation accuracy for Model of [Learning Rate 0.001 | Num Filters 128 | Dropout Rate 0.3 | FilterSize 7]: 0.9659552574157715
Epoch 1/20
36/36 - 2s - loss: 0.3448 - accuracy: 0.8673 - val_loss: 0.1924 - val_accuracy: 0.9309 - 2s/epoch - 43ms/step
Epoch 2/20
36/36 - 1s - loss: 0.1861 - accuracy: 0.9337 - val_loss: 0.1748 - val_accuracy: 0.9360 - 640ms/epoch - 18ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1545 - accuracy: 0.9454 - val_loss: 0.1588 - val_accuracy: 0.9416 - 618ms/epoch - 17ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1446 - accuracy: 0.9495 - val_loss: 0.1281 - val_accuracy: 0.9553 - 590ms/epoch - 16ms/step
Epoch 5/20
36/36 - 1s - loss: 0.1336 - accuracy: 0.9509 - val_loss: 0.1233 - val_accuracy: 0.9573 - 636ms/epoch - 18ms/step
Epoch 6/20
36/36 - 1s - loss: 0.1249 - accuracy: 0.9538 - val_loss: 0.1233 - val_accuracy: 0.9497 - 628ms/epoch - 17ms/step
Epoch 7/20
36/36 - 1s - loss: 0.1209 - accuracy: 0.9538 - val_loss: 0.1278 - val_accuracy: 0.9522 - 643ms/epoch - 18ms/step
Epoch 8/20
36/36 - 1s - loss: 0.1173 - accuracy: 0.9549 - val_loss: 0.1479 - val_accuracy: 0.9405 - 646ms/epoch - 18ms/step
Epoch 9/20
36/36 - 1s - loss: 0.1097 - accuracy: 0.9557 - val_loss: 0.1478 - val_accuracy: 0.9446 - 584ms/epoch - 16ms/step
Epoch 10/20
36/36 - 1s - loss: 0.1076 - accuracy: 0.9566 - val_loss: 0.1103 - val_accuracy: 0.9538 - 727ms/epoch - 20ms/step
Epoch 11/20
36/36 - 1s - loss: 0.1032 - accuracy: 0.9603 - val_loss: 0.1135 - val_accuracy: 0.9578 - 968ms/epoch - 27ms/step
Epoch 12/20
36/36 - 1s - loss: 0.0981 - accuracy: 0.9621 - val_loss: 0.1117 - val_accuracy: 0.9568 - 1s/epoch - 28ms/step
Epoch 13/20
36/36 - 1s - loss: 0.0979 - accuracy: 0.9602 - val_loss: 0.1034 - val_accuracy: 0.9619 - 1s/epoch - 28ms/step
Epoch 14/20
36/36 - 1s - loss: 0.0925 - accuracy: 0.9609 - val_loss: 0.1154 - val_accuracy: 0.9553 - 999ms/epoch - 28ms/step
Epoch 15/20
36/36 - 1s - loss: 0.0912 - accuracy: 0.9621 - val_loss: 0.1017 - val_accuracy: 0.9604 - 831ms/epoch - 23ms/step
Epoch 16/20
36/36 - 1s - loss: 0.0891 - accuracy: 0.9635 - val_loss: 0.0977 - val_accuracy: 0.9624 - 627ms/epoch - 17ms/step
Epoch 17/20
36/36 - 1s - loss: 0.0859 - accuracy: 0.9640 - val_loss: 0.1026 - val_accuracy: 0.9578 - 614ms/epoch - 17ms/step
Epoch 18/20
36/36 - 1s - loss: 0.0866 - accuracy: 0.9655 - val_loss: 0.0969 - val_accuracy: 0.9639 - 639ms/epoch - 18ms/step
Epoch 19/20
36/36 - 1s - loss: 0.0853 - accuracy: 0.9645 - val_loss: 0.1106 - val_accuracy: 0.9548 - 608ms/epoch - 17ms/step
Epoch 20/20
36/36 - 1s - loss: 0.0839 - accuracy: 0.9658 - val_loss: 0.0922 - val_accuracy: 0.9644 - 615ms/epoch - 17ms/step
Validation accuracy for Model of [Learning Rate 0.001 | Num Filters 128 | Dropout Rate 0.5 | FilterSize 3]: 0.9644308686256409
Epoch 1/20
36/36 - 2s - loss: 0.4027 - accuracy: 0.8263 - val_loss: 0.2564 - val_accuracy: 0.9126 - 2s/epoch - 42ms/step
Epoch 2/20
36/36 - 1s - loss: 0.2064 - accuracy: 0.9265 - val_loss: 0.1817 - val_accuracy: 0.9380 - 612ms/epoch - 17ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1601 - accuracy: 0.9439 - val_loss: 0.1563 - val_accuracy: 0.9441 - 608ms/epoch - 17ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1466 - accuracy: 0.9469 - val_loss: 0.1438 - val_accuracy: 0.9472 - 579ms/epoch - 16ms/step
Epoch 5/20
36/36 - 1s - loss: 0.1328 - accuracy: 0.9496 - val_loss: 0.1182 - val_accuracy: 0.9568 - 610ms/epoch - 17ms/step
Epoch 6/20
36/36 - 1s - loss: 0.1256 - accuracy: 0.9522 - val_loss: 0.1275 - val_accuracy: 0.9502 - 583ms/epoch - 16ms/step
Epoch 7/20
36/36 - 1s - loss: 0.1155 - accuracy: 0.9553 - val_loss: 0.1206 - val_accuracy: 0.9548 - 585ms/epoch - 16ms/step
Epoch 8/20
36/36 - 1s - loss: 0.1089 - accuracy: 0.9575 - val_loss: 0.1109 - val_accuracy: 0.9548 - 571ms/epoch - 16ms/step
Epoch 9/20
36/36 - 1s - loss: 0.1028 - accuracy: 0.9595 - val_loss: 0.1168 - val_accuracy: 0.9538 - 636ms/epoch - 18ms/step
Epoch 10/20
36/36 - 1s - loss: 0.0980 - accuracy: 0.9607 - val_loss: 0.1048 - val_accuracy: 0.9583 - 890ms/epoch - 25ms/step
Epoch 11/20
36/36 - 1s - loss: 0.0950 - accuracy: 0.9627 - val_loss: 0.1043 - val_accuracy: 0.9588 - 963ms/epoch - 27ms/step
Epoch 12/20
36/36 - 1s - loss: 0.0933 - accuracy: 0.9634 - val_loss: 0.1154 - val_accuracy: 0.9548 - 944ms/epoch - 26ms/step
Epoch 13/20
36/36 - 1s - loss: 0.0902 - accuracy: 0.9636 - val_loss: 0.1093 - val_accuracy: 0.9568 - 946ms/epoch - 26ms/step
Epoch 14/20
36/36 - 1s - loss: 0.0866 - accuracy: 0.9648 - val_loss: 0.1085 - val_accuracy: 0.9548 - 954ms/epoch - 27ms/step
Epoch 15/20
36/36 - 1s - loss: 0.0812 - accuracy: 0.9663 - val_loss: 0.0965 - val_accuracy: 0.9609 - 624ms/epoch - 17ms/step
Epoch 16/20
36/36 - 1s - loss: 0.0807 - accuracy: 0.9668 - val_loss: 0.1069 - val_accuracy: 0.9563 - 597ms/epoch - 17ms/step
Epoch 17/20
36/36 - 1s - loss: 0.0797 - accuracy: 0.9669 - val_loss: 0.0997 - val_accuracy: 0.9614 - 584ms/epoch - 16ms/step
Epoch 18/20
36/36 - 1s - loss: 0.0787 - accuracy: 0.9684 - val_loss: 0.1032 - val_accuracy: 0.9573 - 584ms/epoch - 16ms/step
Epoch 19/20
36/36 - 1s - loss: 0.0766 - accuracy: 0.9693 - val_loss: 0.0958 - val_accuracy: 0.9614 - 607ms/epoch - 17ms/step
Epoch 20/20
36/36 - 1s - loss: 0.0789 - accuracy: 0.9659 - val_loss: 0.0969 - val_accuracy: 0.9634 - 619ms/epoch - 17ms/step
Validation accuracy for Model of [Learning Rate 0.001 | Num Filters 128 | Dropout Rate 0.5 | FilterSize 5]: 0.9634146094322205
Epoch 1/20
36/36 - 2s - loss: 0.4097 - accuracy: 0.8287 - val_loss: 0.2345 - val_accuracy: 0.9212 - 2s/epoch - 42ms/step
Epoch 2/20
36/36 - 1s - loss: 0.2035 - accuracy: 0.9290 - val_loss: 0.1593 - val_accuracy: 0.9421 - 546ms/epoch - 15ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1558 - accuracy: 0.9443 - val_loss: 0.1213 - val_accuracy: 0.9568 - 514ms/epoch - 14ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1359 - accuracy: 0.9511 - val_loss: 0.1287 - val_accuracy: 0.9527 - 575ms/epoch - 16ms/step
Epoch 5/20
36/36 - 1s - loss: 0.1255 - accuracy: 0.9527 - val_loss: 0.1256 - val_accuracy: 0.9573 - 526ms/epoch - 15ms/step
Epoch 6/20
36/36 - 1s - loss: 0.1147 - accuracy: 0.9561 - val_loss: 0.1280 - val_accuracy: 0.9548 - 562ms/epoch - 16ms/step
Epoch 7/20
36/36 - 1s - loss: 0.1043 - accuracy: 0.9596 - val_loss: 0.1235 - val_accuracy: 0.9517 - 576ms/epoch - 16ms/step
Epoch 8/20
36/36 - 1s - loss: 0.1000 - accuracy: 0.9607 - val_loss: 0.1127 - val_accuracy: 0.9543 - 556ms/epoch - 15ms/step
Epoch 9/20
36/36 - 1s - loss: 0.0925 - accuracy: 0.9629 - val_loss: 0.0954 - val_accuracy: 0.9629 - 572ms/epoch - 16ms/step
Epoch 10/20
36/36 - 1s - loss: 0.0915 - accuracy: 0.9616 - val_loss: 0.0990 - val_accuracy: 0.9624 - 707ms/epoch - 20ms/step
Epoch 11/20
36/36 - 1s - loss: 0.0871 - accuracy: 0.9644 - val_loss: 0.0977 - val_accuracy: 0.9609 - 894ms/epoch - 25ms/step
Epoch 12/20
36/36 - 1s - loss: 0.0860 - accuracy: 0.9653 - val_loss: 0.0876 - val_accuracy: 0.9675 - 872ms/epoch - 24ms/step
Epoch 13/20
36/36 - 1s - loss: 0.0854 - accuracy: 0.9641 - val_loss: 0.0931 - val_accuracy: 0.9665 - 870ms/epoch - 24ms/step
Epoch 14/20
36/36 - 1s - loss: 0.0783 - accuracy: 0.9674 - val_loss: 0.0944 - val_accuracy: 0.9644 - 872ms/epoch - 24ms/step
Epoch 15/20
36/36 - 1s - loss: 0.0779 - accuracy: 0.9682 - val_loss: 0.1030 - val_accuracy: 0.9599 - 869ms/epoch - 24ms/step
Epoch 16/20
36/36 - 1s - loss: 0.0758 - accuracy: 0.9690 - val_loss: 0.0911 - val_accuracy: 0.9649 - 537ms/epoch - 15ms/step
Epoch 17/20
36/36 - 1s - loss: 0.0749 - accuracy: 0.9691 - val_loss: 0.0882 - val_accuracy: 0.9685 - 559ms/epoch - 16ms/step
Epoch 18/20
36/36 - 1s - loss: 0.0751 - accuracy: 0.9693 - val_loss: 0.1081 - val_accuracy: 0.9543 - 514ms/epoch - 14ms/step
Epoch 19/20
36/36 - 1s - loss: 0.0684 - accuracy: 0.9721 - val_loss: 0.1088 - val_accuracy: 0.9533 - 543ms/epoch - 15ms/step
Epoch 20/20
36/36 - 1s - loss: 0.0697 - accuracy: 0.9718 - val_loss: 0.0981 - val_accuracy: 0.9614 - 570ms/epoch - 16ms/step
Validation accuracy for Model of [Learning Rate 0.001 | Num Filters 128 | Dropout Rate 0.5 | FilterSize 7]: 0.9613820910453796
Epoch 1/20
36/36 - 2s - loss: 0.3937 - accuracy: 0.8423 - val_loss: 0.2358 - val_accuracy: 0.9182 - 2s/epoch - 67ms/step
Epoch 2/20
36/36 - 1s - loss: 0.2096 - accuracy: 0.9271 - val_loss: 0.1738 - val_accuracy: 0.9385 - 991ms/epoch - 28ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1696 - accuracy: 0.9380 - val_loss: 0.1717 - val_accuracy: 0.9355 - 889ms/epoch - 25ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1593 - accuracy: 0.9428 - val_loss: 0.1366 - val_accuracy: 0.9477 - 617ms/epoch - 17ms/step
Epoch 5/20
36/36 - 1s - loss: 0.1485 - accuracy: 0.9462 - val_loss: 0.1352 - val_accuracy: 0.9533 - 640ms/epoch - 18ms/step
Epoch 6/20
36/36 - 1s - loss: 0.1390 - accuracy: 0.9475 - val_loss: 0.1368 - val_accuracy: 0.9487 - 626ms/epoch - 17ms/step
Epoch 7/20
36/36 - 1s - loss: 0.1296 - accuracy: 0.9530 - val_loss: 0.1338 - val_accuracy: 0.9507 - 595ms/epoch - 17ms/step
Epoch 8/20
36/36 - 1s - loss: 0.1257 - accuracy: 0.9512 - val_loss: 0.1126 - val_accuracy: 0.9583 - 635ms/epoch - 18ms/step
Epoch 9/20
36/36 - 1s - loss: 0.1246 - accuracy: 0.9552 - val_loss: 0.1113 - val_accuracy: 0.9578 - 583ms/epoch - 16ms/step
Epoch 10/20
36/36 - 1s - loss: 0.1151 - accuracy: 0.9558 - val_loss: 0.1219 - val_accuracy: 0.9497 - 618ms/epoch - 17ms/step
Epoch 11/20
36/36 - 1s - loss: 0.1151 - accuracy: 0.9544 - val_loss: 0.1064 - val_accuracy: 0.9588 - 598ms/epoch - 17ms/step
Epoch 12/20
36/36 - 1s - loss: 0.1109 - accuracy: 0.9555 - val_loss: 0.1070 - val_accuracy: 0.9588 - 613ms/epoch - 17ms/step
Epoch 13/20
36/36 - 1s - loss: 0.1086 - accuracy: 0.9558 - val_loss: 0.1249 - val_accuracy: 0.9482 - 623ms/epoch - 17ms/step
Epoch 14/20
36/36 - 1s - loss: 0.1066 - accuracy: 0.9577 - val_loss: 0.1335 - val_accuracy: 0.9472 - 585ms/epoch - 16ms/step
Epoch 15/20
36/36 - 1s - loss: 0.1041 - accuracy: 0.9574 - val_loss: 0.1073 - val_accuracy: 0.9568 - 613ms/epoch - 17ms/step
Epoch 16/20
36/36 - 1s - loss: 0.1004 - accuracy: 0.9609 - val_loss: 0.1089 - val_accuracy: 0.9553 - 596ms/epoch - 17ms/step
Epoch 17/20
36/36 - 1s - loss: 0.0997 - accuracy: 0.9604 - val_loss: 0.0987 - val_accuracy: 0.9619 - 632ms/epoch - 18ms/step
Epoch 18/20
36/36 - 1s - loss: 0.0995 - accuracy: 0.9590 - val_loss: 0.0999 - val_accuracy: 0.9619 - 593ms/epoch - 16ms/step
Epoch 19/20
36/36 - 1s - loss: 0.0968 - accuracy: 0.9626 - val_loss: 0.1026 - val_accuracy: 0.9583 - 619ms/epoch - 17ms/step
Epoch 20/20
36/36 - 1s - loss: 0.0944 - accuracy: 0.9610 - val_loss: 0.0992 - val_accuracy: 0.9619 - 971ms/epoch - 27ms/step
Validation accuracy for Model of [Learning Rate 0.001 | Num Filters 128 | Dropout Rate 0.7 | FilterSize 3]: 0.9618902206420898
Epoch 1/20
36/36 - 2s - loss: 0.3899 - accuracy: 0.8354 - val_loss: 0.2355 - val_accuracy: 0.9273 - 2s/epoch - 65ms/step
Epoch 2/20
36/36 - 1s - loss: 0.2114 - accuracy: 0.9256 - val_loss: 0.1716 - val_accuracy: 0.9416 - 945ms/epoch - 26ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1705 - accuracy: 0.9379 - val_loss: 0.1771 - val_accuracy: 0.9370 - 655ms/epoch - 18ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1511 - accuracy: 0.9444 - val_loss: 0.1304 - val_accuracy: 0.9497 - 625ms/epoch - 17ms/step
Epoch 5/20
36/36 - 1s - loss: 0.1400 - accuracy: 0.9483 - val_loss: 0.1294 - val_accuracy: 0.9517 - 602ms/epoch - 17ms/step
Epoch 6/20
36/36 - 1s - loss: 0.1293 - accuracy: 0.9525 - val_loss: 0.1187 - val_accuracy: 0.9553 - 584ms/epoch - 16ms/step
Epoch 7/20
36/36 - 1s - loss: 0.1202 - accuracy: 0.9545 - val_loss: 0.1210 - val_accuracy: 0.9527 - 607ms/epoch - 17ms/step
Epoch 8/20
36/36 - 1s - loss: 0.1159 - accuracy: 0.9563 - val_loss: 0.1193 - val_accuracy: 0.9558 - 583ms/epoch - 16ms/step
Epoch 9/20
36/36 - 1s - loss: 0.1105 - accuracy: 0.9577 - val_loss: 0.1214 - val_accuracy: 0.9497 - 605ms/epoch - 17ms/step
Epoch 10/20
36/36 - 1s - loss: 0.1055 - accuracy: 0.9592 - val_loss: 0.0985 - val_accuracy: 0.9624 - 584ms/epoch - 16ms/step
Epoch 11/20
36/36 - 1s - loss: 0.1080 - accuracy: 0.9573 - val_loss: 0.1158 - val_accuracy: 0.9502 - 578ms/epoch - 16ms/step
Epoch 12/20
36/36 - 1s - loss: 0.0979 - accuracy: 0.9604 - val_loss: 0.1068 - val_accuracy: 0.9568 - 562ms/epoch - 16ms/step
Epoch 13/20
36/36 - 1s - loss: 0.0996 - accuracy: 0.9606 - val_loss: 0.1067 - val_accuracy: 0.9563 - 561ms/epoch - 16ms/step
Epoch 14/20
36/36 - 1s - loss: 0.0957 - accuracy: 0.9611 - val_loss: 0.1142 - val_accuracy: 0.9522 - 561ms/epoch - 16ms/step
Epoch 15/20
36/36 - 1s - loss: 0.0933 - accuracy: 0.9638 - val_loss: 0.0949 - val_accuracy: 0.9649 - 599ms/epoch - 17ms/step
Epoch 16/20
36/36 - 1s - loss: 0.0950 - accuracy: 0.9597 - val_loss: 0.1105 - val_accuracy: 0.9502 - 578ms/epoch - 16ms/step
Epoch 17/20
36/36 - 1s - loss: 0.0930 - accuracy: 0.9608 - val_loss: 0.1050 - val_accuracy: 0.9588 - 570ms/epoch - 16ms/step
Epoch 18/20
36/36 - 1s - loss: 0.0878 - accuracy: 0.9651 - val_loss: 0.0981 - val_accuracy: 0.9614 - 615ms/epoch - 17ms/step
Epoch 19/20
36/36 - 1s - loss: 0.0860 - accuracy: 0.9658 - val_loss: 0.0966 - val_accuracy: 0.9593 - 607ms/epoch - 17ms/step
Epoch 20/20
36/36 - 1s - loss: 0.0836 - accuracy: 0.9662 - val_loss: 0.1022 - val_accuracy: 0.9593 - 949ms/epoch - 26ms/step
Validation accuracy for Model of [Learning Rate 0.001 | Num Filters 128 | Dropout Rate 0.7 | FilterSize 5]: 0.9593495726585388
Epoch 1/20
36/36 - 2s - loss: 0.4132 - accuracy: 0.8230 - val_loss: 0.2276 - val_accuracy: 0.9212 - 2s/epoch - 43ms/step
Epoch 2/20
36/36 - 1s - loss: 0.2196 - accuracy: 0.9199 - val_loss: 0.1756 - val_accuracy: 0.9365 - 521ms/epoch - 14ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1682 - accuracy: 0.9397 - val_loss: 0.1351 - val_accuracy: 0.9492 - 522ms/epoch - 14ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1520 - accuracy: 0.9448 - val_loss: 0.1653 - val_accuracy: 0.9355 - 553ms/epoch - 15ms/step
Epoch 5/20
36/36 - 1s - loss: 0.1423 - accuracy: 0.9474 - val_loss: 0.1215 - val_accuracy: 0.9558 - 548ms/epoch - 15ms/step
Epoch 6/20
36/36 - 1s - loss: 0.1258 - accuracy: 0.9531 - val_loss: 0.1087 - val_accuracy: 0.9578 - 556ms/epoch - 15ms/step
Epoch 7/20
36/36 - 1s - loss: 0.1177 - accuracy: 0.9555 - val_loss: 0.1009 - val_accuracy: 0.9599 - 513ms/epoch - 14ms/step
Epoch 8/20
36/36 - 1s - loss: 0.1143 - accuracy: 0.9567 - val_loss: 0.1082 - val_accuracy: 0.9573 - 572ms/epoch - 16ms/step
Epoch 9/20
36/36 - 1s - loss: 0.1087 - accuracy: 0.9570 - val_loss: 0.1076 - val_accuracy: 0.9563 - 532ms/epoch - 15ms/step
Epoch 10/20
36/36 - 1s - loss: 0.1034 - accuracy: 0.9586 - val_loss: 0.1123 - val_accuracy: 0.9543 - 543ms/epoch - 15ms/step
Epoch 11/20
36/36 - 1s - loss: 0.0999 - accuracy: 0.9607 - val_loss: 0.1165 - val_accuracy: 0.9517 - 630ms/epoch - 17ms/step
Epoch 12/20
36/36 - 1s - loss: 0.0980 - accuracy: 0.9612 - val_loss: 0.1152 - val_accuracy: 0.9507 - 877ms/epoch - 24ms/step
Epoch 13/20
36/36 - 1s - loss: 0.0922 - accuracy: 0.9628 - val_loss: 0.0997 - val_accuracy: 0.9604 - 862ms/epoch - 24ms/step
Epoch 14/20
36/36 - 1s - loss: 0.0908 - accuracy: 0.9638 - val_loss: 0.1221 - val_accuracy: 0.9466 - 844ms/epoch - 23ms/step
Epoch 15/20
36/36 - 1s - loss: 0.0866 - accuracy: 0.9648 - val_loss: 0.0987 - val_accuracy: 0.9634 - 845ms/epoch - 23ms/step
Epoch 16/20
36/36 - 1s - loss: 0.0880 - accuracy: 0.9653 - val_loss: 0.0960 - val_accuracy: 0.9614 - 860ms/epoch - 24ms/step
Epoch 17/20
36/36 - 1s - loss: 0.0801 - accuracy: 0.9673 - val_loss: 0.1007 - val_accuracy: 0.9619 - 607ms/epoch - 17ms/step
Epoch 18/20
36/36 - 1s - loss: 0.0838 - accuracy: 0.9656 - val_loss: 0.1038 - val_accuracy: 0.9558 - 523ms/epoch - 15ms/step
Epoch 19/20
36/36 - 1s - loss: 0.0803 - accuracy: 0.9678 - val_loss: 0.0990 - val_accuracy: 0.9558 - 511ms/epoch - 14ms/step
Epoch 20/20
36/36 - 1s - loss: 0.0803 - accuracy: 0.9677 - val_loss: 0.0940 - val_accuracy: 0.9639 - 528ms/epoch - 15ms/step
Validation accuracy for Model of [Learning Rate 0.001 | Num Filters 128 | Dropout Rate 0.7 | FilterSize 7]: 0.9639227390289307
Epoch 1/20
36/36 - 2s - loss: 0.2589 - accuracy: 0.8929 - val_loss: 0.1716 - val_accuracy: 0.9360 - 2s/epoch - 49ms/step
Epoch 2/20
36/36 - 0s - loss: 0.1376 - accuracy: 0.9454 - val_loss: 0.1164 - val_accuracy: 0.9583 - 434ms/epoch - 12ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1157 - accuracy: 0.9523 - val_loss: 0.1039 - val_accuracy: 0.9604 - 451ms/epoch - 13ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1129 - accuracy: 0.9537 - val_loss: 0.0864 - val_accuracy: 0.9695 - 449ms/epoch - 12ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1015 - accuracy: 0.9597 - val_loss: 0.1105 - val_accuracy: 0.9543 - 416ms/epoch - 12ms/step
Epoch 6/20
36/36 - 0s - loss: 0.0914 - accuracy: 0.9613 - val_loss: 0.0992 - val_accuracy: 0.9634 - 429ms/epoch - 12ms/step
Epoch 7/20
36/36 - 0s - loss: 0.0804 - accuracy: 0.9664 - val_loss: 0.0828 - val_accuracy: 0.9695 - 466ms/epoch - 13ms/step
Epoch 8/20
36/36 - 0s - loss: 0.0860 - accuracy: 0.9649 - val_loss: 0.0970 - val_accuracy: 0.9604 - 473ms/epoch - 13ms/step
Epoch 9/20
36/36 - 0s - loss: 0.0761 - accuracy: 0.9691 - val_loss: 0.0985 - val_accuracy: 0.9654 - 284ms/epoch - 8ms/step
Epoch 10/20
36/36 - 0s - loss: 0.0743 - accuracy: 0.9699 - val_loss: 0.1009 - val_accuracy: 0.9573 - 288ms/epoch - 8ms/step
Epoch 11/20
36/36 - 0s - loss: 0.0752 - accuracy: 0.9688 - val_loss: 0.1235 - val_accuracy: 0.9583 - 280ms/epoch - 8ms/step
Epoch 12/20
36/36 - 0s - loss: 0.0731 - accuracy: 0.9707 - val_loss: 0.0813 - val_accuracy: 0.9746 - 288ms/epoch - 8ms/step
Epoch 13/20
36/36 - 0s - loss: 0.0645 - accuracy: 0.9751 - val_loss: 0.0792 - val_accuracy: 0.9705 - 279ms/epoch - 8ms/step
Epoch 14/20
36/36 - 0s - loss: 0.0612 - accuracy: 0.9754 - val_loss: 0.0875 - val_accuracy: 0.9705 - 294ms/epoch - 8ms/step
Epoch 15/20
36/36 - 0s - loss: 0.0708 - accuracy: 0.9728 - val_loss: 0.0846 - val_accuracy: 0.9705 - 296ms/epoch - 8ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0694 - accuracy: 0.9724 - val_loss: 0.0885 - val_accuracy: 0.9639 - 277ms/epoch - 8ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0652 - accuracy: 0.9733 - val_loss: 0.0770 - val_accuracy: 0.9736 - 298ms/epoch - 8ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0641 - accuracy: 0.9750 - val_loss: 0.0840 - val_accuracy: 0.9746 - 286ms/epoch - 8ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0600 - accuracy: 0.9772 - val_loss: 0.0870 - val_accuracy: 0.9710 - 338ms/epoch - 9ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0586 - accuracy: 0.9763 - val_loss: 0.1006 - val_accuracy: 0.9705 - 278ms/epoch - 8ms/step
Validation accuracy for Model of [Learning Rate 0.01 | Num Filters 32 | Dropout Rate 0.3 | FilterSize 3]: 0.9705284833908081
Epoch 1/20
36/36 - 1s - loss: 0.2832 - accuracy: 0.8812 - val_loss: 0.1524 - val_accuracy: 0.9416 - 1s/epoch - 36ms/step
Epoch 2/20
36/36 - 0s - loss: 0.1367 - accuracy: 0.9469 - val_loss: 0.1263 - val_accuracy: 0.9527 - 294ms/epoch - 8ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1115 - accuracy: 0.9559 - val_loss: 0.1369 - val_accuracy: 0.9492 - 276ms/epoch - 8ms/step
Epoch 4/20
36/36 - 0s - loss: 0.0988 - accuracy: 0.9606 - val_loss: 0.1184 - val_accuracy: 0.9543 - 298ms/epoch - 8ms/step
Epoch 5/20
36/36 - 0s - loss: 0.0887 - accuracy: 0.9637 - val_loss: 0.0930 - val_accuracy: 0.9660 - 281ms/epoch - 8ms/step
Epoch 6/20
36/36 - 0s - loss: 0.0787 - accuracy: 0.9668 - val_loss: 0.0930 - val_accuracy: 0.9619 - 290ms/epoch - 8ms/step
Epoch 7/20
36/36 - 0s - loss: 0.0808 - accuracy: 0.9658 - val_loss: 0.0803 - val_accuracy: 0.9721 - 285ms/epoch - 8ms/step
Epoch 8/20
36/36 - 0s - loss: 0.0782 - accuracy: 0.9693 - val_loss: 0.1051 - val_accuracy: 0.9578 - 278ms/epoch - 8ms/step
Epoch 9/20
36/36 - 0s - loss: 0.0705 - accuracy: 0.9704 - val_loss: 0.0925 - val_accuracy: 0.9649 - 281ms/epoch - 8ms/step
Epoch 10/20
36/36 - 0s - loss: 0.0705 - accuracy: 0.9707 - val_loss: 0.0828 - val_accuracy: 0.9741 - 289ms/epoch - 8ms/step
Epoch 11/20
36/36 - 0s - loss: 0.0670 - accuracy: 0.9727 - val_loss: 0.1256 - val_accuracy: 0.9461 - 298ms/epoch - 8ms/step
Epoch 12/20
36/36 - 0s - loss: 0.0655 - accuracy: 0.9739 - val_loss: 0.0973 - val_accuracy: 0.9634 - 287ms/epoch - 8ms/step
Epoch 13/20
36/36 - 0s - loss: 0.0639 - accuracy: 0.9758 - val_loss: 0.0877 - val_accuracy: 0.9680 - 265ms/epoch - 7ms/step
Epoch 14/20
36/36 - 0s - loss: 0.0619 - accuracy: 0.9759 - val_loss: 0.0902 - val_accuracy: 0.9721 - 263ms/epoch - 7ms/step
Epoch 15/20
36/36 - 0s - loss: 0.0638 - accuracy: 0.9759 - val_loss: 0.0815 - val_accuracy: 0.9776 - 299ms/epoch - 8ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0659 - accuracy: 0.9740 - val_loss: 0.0942 - val_accuracy: 0.9680 - 276ms/epoch - 8ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0546 - accuracy: 0.9783 - val_loss: 0.0807 - val_accuracy: 0.9756 - 331ms/epoch - 9ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0510 - accuracy: 0.9807 - val_loss: 0.0936 - val_accuracy: 0.9715 - 338ms/epoch - 9ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0647 - accuracy: 0.9776 - val_loss: 0.0930 - val_accuracy: 0.9766 - 398ms/epoch - 11ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0538 - accuracy: 0.9780 - val_loss: 0.0912 - val_accuracy: 0.9726 - 433ms/epoch - 12ms/step
Validation accuracy for Model of [Learning Rate 0.01 | Num Filters 32 | Dropout Rate 0.3 | FilterSize 5]: 0.9725610017776489
Epoch 1/20
36/36 - 2s - loss: 0.2549 - accuracy: 0.8934 - val_loss: 0.1255 - val_accuracy: 0.9517 - 2s/epoch - 52ms/step
Epoch 2/20
36/36 - 0s - loss: 0.1162 - accuracy: 0.9560 - val_loss: 0.1049 - val_accuracy: 0.9578 - 435ms/epoch - 12ms/step
Epoch 3/20
36/36 - 0s - loss: 0.0987 - accuracy: 0.9600 - val_loss: 0.1091 - val_accuracy: 0.9568 - 436ms/epoch - 12ms/step
Epoch 4/20
36/36 - 0s - loss: 0.0921 - accuracy: 0.9621 - val_loss: 0.0950 - val_accuracy: 0.9649 - 464ms/epoch - 13ms/step
Epoch 5/20
36/36 - 0s - loss: 0.0840 - accuracy: 0.9656 - val_loss: 0.1284 - val_accuracy: 0.9477 - 401ms/epoch - 11ms/step
Epoch 6/20
36/36 - 0s - loss: 0.0824 - accuracy: 0.9654 - val_loss: 0.1145 - val_accuracy: 0.9548 - 279ms/epoch - 8ms/step
Epoch 7/20
36/36 - 0s - loss: 0.0724 - accuracy: 0.9705 - val_loss: 0.0901 - val_accuracy: 0.9715 - 278ms/epoch - 8ms/step
Epoch 8/20
36/36 - 0s - loss: 0.0727 - accuracy: 0.9705 - val_loss: 0.0912 - val_accuracy: 0.9731 - 278ms/epoch - 8ms/step
Epoch 9/20
36/36 - 0s - loss: 0.0629 - accuracy: 0.9747 - val_loss: 0.0970 - val_accuracy: 0.9609 - 271ms/epoch - 8ms/step
Epoch 10/20
36/36 - 0s - loss: 0.0617 - accuracy: 0.9744 - val_loss: 0.1033 - val_accuracy: 0.9660 - 279ms/epoch - 8ms/step
Epoch 11/20
36/36 - 0s - loss: 0.0682 - accuracy: 0.9739 - val_loss: 0.0961 - val_accuracy: 0.9644 - 274ms/epoch - 8ms/step
Epoch 12/20
36/36 - 0s - loss: 0.0614 - accuracy: 0.9761 - val_loss: 0.1031 - val_accuracy: 0.9670 - 314ms/epoch - 9ms/step
Epoch 13/20
36/36 - 0s - loss: 0.0681 - accuracy: 0.9728 - val_loss: 0.1066 - val_accuracy: 0.9573 - 320ms/epoch - 9ms/step
Epoch 14/20
36/36 - 0s - loss: 0.0641 - accuracy: 0.9748 - val_loss: 0.0880 - val_accuracy: 0.9705 - 282ms/epoch - 8ms/step
Epoch 15/20
36/36 - 0s - loss: 0.0558 - accuracy: 0.9793 - val_loss: 0.1072 - val_accuracy: 0.9624 - 275ms/epoch - 8ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0521 - accuracy: 0.9804 - val_loss: 0.0976 - val_accuracy: 0.9690 - 317ms/epoch - 9ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0544 - accuracy: 0.9785 - val_loss: 0.0940 - val_accuracy: 0.9715 - 289ms/epoch - 8ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0527 - accuracy: 0.9778 - val_loss: 0.0946 - val_accuracy: 0.9670 - 267ms/epoch - 7ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0551 - accuracy: 0.9787 - val_loss: 0.0921 - val_accuracy: 0.9766 - 268ms/epoch - 7ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0529 - accuracy: 0.9800 - val_loss: 0.0869 - val_accuracy: 0.9741 - 286ms/epoch - 8ms/step
Validation accuracy for Model of [Learning Rate 0.01 | Num Filters 32 | Dropout Rate 0.3 | FilterSize 7]: 0.9740853905677795
Epoch 1/20
36/36 - 1s - loss: 0.2476 - accuracy: 0.9012 - val_loss: 0.1359 - val_accuracy: 0.9538 - 1s/epoch - 35ms/step
Epoch 2/20
36/36 - 0s - loss: 0.1480 - accuracy: 0.9445 - val_loss: 0.1480 - val_accuracy: 0.9466 - 334ms/epoch - 9ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1287 - accuracy: 0.9519 - val_loss: 0.1241 - val_accuracy: 0.9487 - 281ms/epoch - 8ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1101 - accuracy: 0.9556 - val_loss: 0.1200 - val_accuracy: 0.9517 - 383ms/epoch - 11ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1006 - accuracy: 0.9586 - val_loss: 0.0878 - val_accuracy: 0.9665 - 481ms/epoch - 13ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1032 - accuracy: 0.9578 - val_loss: 0.0964 - val_accuracy: 0.9644 - 425ms/epoch - 12ms/step
Epoch 7/20
36/36 - 0s - loss: 0.0995 - accuracy: 0.9590 - val_loss: 0.0979 - val_accuracy: 0.9619 - 425ms/epoch - 12ms/step
Epoch 8/20
36/36 - 0s - loss: 0.0947 - accuracy: 0.9622 - val_loss: 0.1049 - val_accuracy: 0.9573 - 465ms/epoch - 13ms/step
Epoch 9/20
36/36 - 0s - loss: 0.0874 - accuracy: 0.9636 - val_loss: 0.0906 - val_accuracy: 0.9700 - 469ms/epoch - 13ms/step
Epoch 10/20
36/36 - 0s - loss: 0.0897 - accuracy: 0.9638 - val_loss: 0.1165 - val_accuracy: 0.9548 - 488ms/epoch - 14ms/step
Epoch 11/20
36/36 - 0s - loss: 0.0840 - accuracy: 0.9663 - val_loss: 0.0850 - val_accuracy: 0.9700 - 474ms/epoch - 13ms/step
Epoch 12/20
36/36 - 0s - loss: 0.0826 - accuracy: 0.9671 - val_loss: 0.0852 - val_accuracy: 0.9726 - 449ms/epoch - 12ms/step
Epoch 13/20
36/36 - 0s - loss: 0.0779 - accuracy: 0.9673 - val_loss: 0.0981 - val_accuracy: 0.9660 - 469ms/epoch - 13ms/step
Epoch 14/20
36/36 - 0s - loss: 0.0772 - accuracy: 0.9681 - val_loss: 0.0939 - val_accuracy: 0.9710 - 417ms/epoch - 12ms/step
Epoch 15/20
36/36 - 0s - loss: 0.0759 - accuracy: 0.9683 - val_loss: 0.0902 - val_accuracy: 0.9700 - 278ms/epoch - 8ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0791 - accuracy: 0.9670 - val_loss: 0.0978 - val_accuracy: 0.9690 - 277ms/epoch - 8ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0726 - accuracy: 0.9711 - val_loss: 0.0993 - val_accuracy: 0.9670 - 282ms/epoch - 8ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0697 - accuracy: 0.9718 - val_loss: 0.0956 - val_accuracy: 0.9685 - 299ms/epoch - 8ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0706 - accuracy: 0.9697 - val_loss: 0.0873 - val_accuracy: 0.9705 - 298ms/epoch - 8ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0757 - accuracy: 0.9699 - val_loss: 0.1016 - val_accuracy: 0.9639 - 287ms/epoch - 8ms/step
Validation accuracy for Model of [Learning Rate 0.01 | Num Filters 32 | Dropout Rate 0.5 | FilterSize 3]: 0.9639227390289307
Epoch 1/20
36/36 - 1s - loss: 0.2639 - accuracy: 0.8904 - val_loss: 0.1391 - val_accuracy: 0.9451 - 1s/epoch - 35ms/step
Epoch 2/20
36/36 - 0s - loss: 0.1404 - accuracy: 0.9454 - val_loss: 0.1027 - val_accuracy: 0.9588 - 284ms/epoch - 8ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1171 - accuracy: 0.9540 - val_loss: 0.1108 - val_accuracy: 0.9604 - 281ms/epoch - 8ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1158 - accuracy: 0.9522 - val_loss: 0.1001 - val_accuracy: 0.9563 - 275ms/epoch - 8ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1027 - accuracy: 0.9580 - val_loss: 0.0992 - val_accuracy: 0.9604 - 291ms/epoch - 8ms/step
Epoch 6/20
36/36 - 0s - loss: 0.0963 - accuracy: 0.9588 - val_loss: 0.0995 - val_accuracy: 0.9604 - 270ms/epoch - 8ms/step
Epoch 7/20
36/36 - 0s - loss: 0.0927 - accuracy: 0.9615 - val_loss: 0.0969 - val_accuracy: 0.9593 - 290ms/epoch - 8ms/step
Epoch 8/20
36/36 - 0s - loss: 0.0923 - accuracy: 0.9626 - val_loss: 0.1167 - val_accuracy: 0.9507 - 292ms/epoch - 8ms/step
Epoch 9/20
36/36 - 0s - loss: 0.0873 - accuracy: 0.9632 - val_loss: 0.0960 - val_accuracy: 0.9654 - 278ms/epoch - 8ms/step
Epoch 10/20
36/36 - 0s - loss: 0.0881 - accuracy: 0.9616 - val_loss: 0.1106 - val_accuracy: 0.9497 - 267ms/epoch - 7ms/step
Epoch 11/20
36/36 - 0s - loss: 0.0800 - accuracy: 0.9674 - val_loss: 0.0927 - val_accuracy: 0.9644 - 270ms/epoch - 8ms/step
Epoch 12/20
36/36 - 0s - loss: 0.0814 - accuracy: 0.9676 - val_loss: 0.0953 - val_accuracy: 0.9649 - 297ms/epoch - 8ms/step
Epoch 13/20
36/36 - 0s - loss: 0.0782 - accuracy: 0.9676 - val_loss: 0.0930 - val_accuracy: 0.9665 - 277ms/epoch - 8ms/step
Epoch 14/20
36/36 - 0s - loss: 0.0741 - accuracy: 0.9691 - val_loss: 0.0918 - val_accuracy: 0.9731 - 267ms/epoch - 7ms/step
Epoch 15/20
36/36 - 0s - loss: 0.0731 - accuracy: 0.9699 - val_loss: 0.0908 - val_accuracy: 0.9685 - 306ms/epoch - 8ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0691 - accuracy: 0.9731 - val_loss: 0.0903 - val_accuracy: 0.9695 - 312ms/epoch - 9ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0715 - accuracy: 0.9703 - val_loss: 0.0893 - val_accuracy: 0.9736 - 274ms/epoch - 8ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0690 - accuracy: 0.9727 - val_loss: 0.0936 - val_accuracy: 0.9741 - 278ms/epoch - 8ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0727 - accuracy: 0.9685 - val_loss: 0.0936 - val_accuracy: 0.9639 - 294ms/epoch - 8ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0702 - accuracy: 0.9724 - val_loss: 0.1049 - val_accuracy: 0.9558 - 281ms/epoch - 8ms/step
Validation accuracy for Model of [Learning Rate 0.01 | Num Filters 32 | Dropout Rate 0.5 | FilterSize 5]: 0.9557926654815674
Epoch 1/20
36/36 - 1s - loss: 0.2773 - accuracy: 0.8814 - val_loss: 0.1592 - val_accuracy: 0.9421 - 1s/epoch - 37ms/step
Epoch 2/20
36/36 - 0s - loss: 0.1434 - accuracy: 0.9464 - val_loss: 0.1312 - val_accuracy: 0.9533 - 468ms/epoch - 13ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1182 - accuracy: 0.9539 - val_loss: 0.1312 - val_accuracy: 0.9507 - 429ms/epoch - 12ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1069 - accuracy: 0.9583 - val_loss: 0.1183 - val_accuracy: 0.9563 - 443ms/epoch - 12ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1037 - accuracy: 0.9580 - val_loss: 0.1168 - val_accuracy: 0.9553 - 425ms/epoch - 12ms/step
Epoch 6/20
36/36 - 0s - loss: 0.0927 - accuracy: 0.9635 - val_loss: 0.0923 - val_accuracy: 0.9665 - 453ms/epoch - 13ms/step
Epoch 7/20
36/36 - 0s - loss: 0.0894 - accuracy: 0.9647 - val_loss: 0.1038 - val_accuracy: 0.9583 - 461ms/epoch - 13ms/step
Epoch 8/20
36/36 - 0s - loss: 0.0905 - accuracy: 0.9641 - val_loss: 0.0997 - val_accuracy: 0.9665 - 426ms/epoch - 12ms/step
Epoch 9/20
36/36 - 0s - loss: 0.0836 - accuracy: 0.9665 - val_loss: 0.1027 - val_accuracy: 0.9614 - 429ms/epoch - 12ms/step
Epoch 10/20
36/36 - 0s - loss: 0.0841 - accuracy: 0.9649 - val_loss: 0.0948 - val_accuracy: 0.9665 - 451ms/epoch - 13ms/step
Epoch 11/20
36/36 - 0s - loss: 0.0773 - accuracy: 0.9690 - val_loss: 0.0922 - val_accuracy: 0.9690 - 436ms/epoch - 12ms/step
Epoch 12/20
36/36 - 0s - loss: 0.0754 - accuracy: 0.9702 - val_loss: 0.0875 - val_accuracy: 0.9665 - 271ms/epoch - 8ms/step
Epoch 13/20
36/36 - 0s - loss: 0.0739 - accuracy: 0.9689 - val_loss: 0.0947 - val_accuracy: 0.9670 - 269ms/epoch - 7ms/step
Epoch 14/20
36/36 - 0s - loss: 0.0743 - accuracy: 0.9707 - val_loss: 0.0990 - val_accuracy: 0.9649 - 269ms/epoch - 7ms/step
Epoch 15/20
36/36 - 0s - loss: 0.0704 - accuracy: 0.9718 - val_loss: 0.1071 - val_accuracy: 0.9527 - 270ms/epoch - 7ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0712 - accuracy: 0.9711 - val_loss: 0.0923 - val_accuracy: 0.9695 - 268ms/epoch - 7ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0730 - accuracy: 0.9699 - val_loss: 0.0982 - val_accuracy: 0.9690 - 308ms/epoch - 9ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0652 - accuracy: 0.9739 - val_loss: 0.0872 - val_accuracy: 0.9756 - 270ms/epoch - 7ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0642 - accuracy: 0.9738 - val_loss: 0.0927 - val_accuracy: 0.9705 - 272ms/epoch - 8ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0629 - accuracy: 0.9733 - val_loss: 0.0897 - val_accuracy: 0.9741 - 258ms/epoch - 7ms/step
Validation accuracy for Model of [Learning Rate 0.01 | Num Filters 32 | Dropout Rate 0.5 | FilterSize 7]: 0.9740853905677795
Epoch 1/20
36/36 - 1s - loss: 0.2693 - accuracy: 0.8876 - val_loss: 0.1379 - val_accuracy: 0.9507 - 1s/epoch - 35ms/step
Epoch 2/20
36/36 - 0s - loss: 0.1687 - accuracy: 0.9378 - val_loss: 0.1484 - val_accuracy: 0.9487 - 298ms/epoch - 8ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1453 - accuracy: 0.9440 - val_loss: 0.1227 - val_accuracy: 0.9512 - 280ms/epoch - 8ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1334 - accuracy: 0.9472 - val_loss: 0.1290 - val_accuracy: 0.9482 - 286ms/epoch - 8ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1277 - accuracy: 0.9471 - val_loss: 0.1093 - val_accuracy: 0.9548 - 309ms/epoch - 9ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1223 - accuracy: 0.9506 - val_loss: 0.1176 - val_accuracy: 0.9533 - 288ms/epoch - 8ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1219 - accuracy: 0.9512 - val_loss: 0.1122 - val_accuracy: 0.9553 - 289ms/epoch - 8ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1243 - accuracy: 0.9478 - val_loss: 0.1151 - val_accuracy: 0.9512 - 301ms/epoch - 8ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1121 - accuracy: 0.9553 - val_loss: 0.1080 - val_accuracy: 0.9593 - 289ms/epoch - 8ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1031 - accuracy: 0.9581 - val_loss: 0.1004 - val_accuracy: 0.9593 - 276ms/epoch - 8ms/step
Epoch 11/20
36/36 - 0s - loss: 0.1052 - accuracy: 0.9569 - val_loss: 0.1033 - val_accuracy: 0.9644 - 287ms/epoch - 8ms/step
Epoch 12/20
36/36 - 0s - loss: 0.1033 - accuracy: 0.9589 - val_loss: 0.1051 - val_accuracy: 0.9593 - 303ms/epoch - 8ms/step
Epoch 13/20
36/36 - 0s - loss: 0.0996 - accuracy: 0.9597 - val_loss: 0.0941 - val_accuracy: 0.9639 - 299ms/epoch - 8ms/step
Epoch 14/20
36/36 - 0s - loss: 0.0997 - accuracy: 0.9607 - val_loss: 0.1092 - val_accuracy: 0.9588 - 285ms/epoch - 8ms/step
Epoch 15/20
36/36 - 0s - loss: 0.0994 - accuracy: 0.9591 - val_loss: 0.1095 - val_accuracy: 0.9588 - 281ms/epoch - 8ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0960 - accuracy: 0.9596 - val_loss: 0.0954 - val_accuracy: 0.9644 - 326ms/epoch - 9ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0991 - accuracy: 0.9591 - val_loss: 0.1032 - val_accuracy: 0.9604 - 292ms/epoch - 8ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0950 - accuracy: 0.9607 - val_loss: 0.1153 - val_accuracy: 0.9573 - 297ms/epoch - 8ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0919 - accuracy: 0.9622 - val_loss: 0.0906 - val_accuracy: 0.9700 - 290ms/epoch - 8ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0948 - accuracy: 0.9629 - val_loss: 0.1006 - val_accuracy: 0.9614 - 284ms/epoch - 8ms/step
Validation accuracy for Model of [Learning Rate 0.01 | Num Filters 32 | Dropout Rate 0.7 | FilterSize 3]: 0.9613820910453796
Epoch 1/20
36/36 - 1s - loss: 0.3179 - accuracy: 0.8578 - val_loss: 0.1520 - val_accuracy: 0.9451 - 1s/epoch - 34ms/step
Epoch 2/20
36/36 - 0s - loss: 0.1584 - accuracy: 0.9401 - val_loss: 0.1358 - val_accuracy: 0.9507 - 278ms/epoch - 8ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1416 - accuracy: 0.9466 - val_loss: 0.1217 - val_accuracy: 0.9517 - 296ms/epoch - 8ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1307 - accuracy: 0.9473 - val_loss: 0.1123 - val_accuracy: 0.9558 - 283ms/epoch - 8ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1185 - accuracy: 0.9542 - val_loss: 0.1190 - val_accuracy: 0.9492 - 292ms/epoch - 8ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1122 - accuracy: 0.9569 - val_loss: 0.1144 - val_accuracy: 0.9533 - 294ms/epoch - 8ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1113 - accuracy: 0.9546 - val_loss: 0.1136 - val_accuracy: 0.9517 - 327ms/epoch - 9ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1172 - accuracy: 0.9535 - val_loss: 0.1123 - val_accuracy: 0.9522 - 287ms/epoch - 8ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1052 - accuracy: 0.9578 - val_loss: 0.1052 - val_accuracy: 0.9609 - 325ms/epoch - 9ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1011 - accuracy: 0.9563 - val_loss: 0.1016 - val_accuracy: 0.9634 - 324ms/epoch - 9ms/step
Epoch 11/20
36/36 - 0s - loss: 0.1008 - accuracy: 0.9608 - val_loss: 0.1058 - val_accuracy: 0.9543 - 298ms/epoch - 8ms/step
Epoch 12/20
36/36 - 0s - loss: 0.1025 - accuracy: 0.9575 - val_loss: 0.0966 - val_accuracy: 0.9639 - 320ms/epoch - 9ms/step
Epoch 13/20
36/36 - 0s - loss: 0.1004 - accuracy: 0.9601 - val_loss: 0.1013 - val_accuracy: 0.9573 - 282ms/epoch - 8ms/step
Epoch 14/20
36/36 - 0s - loss: 0.1020 - accuracy: 0.9590 - val_loss: 0.0930 - val_accuracy: 0.9690 - 300ms/epoch - 8ms/step
Epoch 15/20
36/36 - 0s - loss: 0.1027 - accuracy: 0.9612 - val_loss: 0.1298 - val_accuracy: 0.9436 - 292ms/epoch - 8ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0953 - accuracy: 0.9613 - val_loss: 0.1185 - val_accuracy: 0.9502 - 289ms/epoch - 8ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0918 - accuracy: 0.9614 - val_loss: 0.1068 - val_accuracy: 0.9573 - 276ms/epoch - 8ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0885 - accuracy: 0.9635 - val_loss: 0.0928 - val_accuracy: 0.9634 - 273ms/epoch - 8ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0892 - accuracy: 0.9629 - val_loss: 0.0951 - val_accuracy: 0.9675 - 284ms/epoch - 8ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0890 - accuracy: 0.9633 - val_loss: 0.1032 - val_accuracy: 0.9629 - 345ms/epoch - 10ms/step
Validation accuracy for Model of [Learning Rate 0.01 | Num Filters 32 | Dropout Rate 0.7 | FilterSize 5]: 0.9629064798355103
Epoch 1/20
36/36 - 1s - loss: 0.2891 - accuracy: 0.8732 - val_loss: 0.1344 - val_accuracy: 0.9558 - 1s/epoch - 34ms/step
Epoch 2/20
36/36 - 0s - loss: 0.1482 - accuracy: 0.9428 - val_loss: 0.1203 - val_accuracy: 0.9558 - 266ms/epoch - 7ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1289 - accuracy: 0.9487 - val_loss: 0.1051 - val_accuracy: 0.9578 - 279ms/epoch - 8ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1163 - accuracy: 0.9544 - val_loss: 0.1030 - val_accuracy: 0.9654 - 412ms/epoch - 11ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1051 - accuracy: 0.9586 - val_loss: 0.1256 - val_accuracy: 0.9477 - 435ms/epoch - 12ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1046 - accuracy: 0.9589 - val_loss: 0.1040 - val_accuracy: 0.9573 - 441ms/epoch - 12ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1068 - accuracy: 0.9579 - val_loss: 0.1203 - val_accuracy: 0.9466 - 437ms/epoch - 12ms/step
Epoch 8/20
36/36 - 0s - loss: 0.0965 - accuracy: 0.9618 - val_loss: 0.1093 - val_accuracy: 0.9553 - 476ms/epoch - 13ms/step
Epoch 9/20
36/36 - 0s - loss: 0.0930 - accuracy: 0.9641 - val_loss: 0.1033 - val_accuracy: 0.9588 - 449ms/epoch - 12ms/step
Epoch 10/20
36/36 - 0s - loss: 0.0883 - accuracy: 0.9635 - val_loss: 0.1179 - val_accuracy: 0.9527 - 488ms/epoch - 14ms/step
Epoch 11/20
36/36 - 0s - loss: 0.0866 - accuracy: 0.9638 - val_loss: 0.1056 - val_accuracy: 0.9609 - 493ms/epoch - 14ms/step
Epoch 12/20
36/36 - 0s - loss: 0.0881 - accuracy: 0.9641 - val_loss: 0.1070 - val_accuracy: 0.9604 - 483ms/epoch - 13ms/step
Epoch 13/20
36/36 - 0s - loss: 0.0899 - accuracy: 0.9644 - val_loss: 0.0946 - val_accuracy: 0.9619 - 479ms/epoch - 13ms/step
Epoch 14/20
36/36 - 0s - loss: 0.0825 - accuracy: 0.9684 - val_loss: 0.0996 - val_accuracy: 0.9639 - 304ms/epoch - 8ms/step
Epoch 15/20
36/36 - 0s - loss: 0.0850 - accuracy: 0.9626 - val_loss: 0.1194 - val_accuracy: 0.9568 - 293ms/epoch - 8ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0826 - accuracy: 0.9651 - val_loss: 0.1029 - val_accuracy: 0.9578 - 278ms/epoch - 8ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0819 - accuracy: 0.9664 - val_loss: 0.1333 - val_accuracy: 0.9461 - 264ms/epoch - 7ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0881 - accuracy: 0.9639 - val_loss: 0.1021 - val_accuracy: 0.9527 - 270ms/epoch - 8ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0818 - accuracy: 0.9645 - val_loss: 0.1058 - val_accuracy: 0.9517 - 297ms/epoch - 8ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0819 - accuracy: 0.9666 - val_loss: 0.1089 - val_accuracy: 0.9583 - 289ms/epoch - 8ms/step
Validation accuracy for Model of [Learning Rate 0.01 | Num Filters 32 | Dropout Rate 0.7 | FilterSize 7]: 0.9583333134651184
Epoch 1/20
36/36 - 1s - loss: 0.2336 - accuracy: 0.9016 - val_loss: 0.1320 - val_accuracy: 0.9568 - 1s/epoch - 38ms/step
Epoch 2/20
36/36 - 0s - loss: 0.1308 - accuracy: 0.9479 - val_loss: 0.1168 - val_accuracy: 0.9497 - 388ms/epoch - 11ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1160 - accuracy: 0.9527 - val_loss: 0.0967 - val_accuracy: 0.9634 - 405ms/epoch - 11ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1040 - accuracy: 0.9555 - val_loss: 0.0963 - val_accuracy: 0.9685 - 416ms/epoch - 12ms/step
Epoch 5/20
36/36 - 0s - loss: 0.0915 - accuracy: 0.9614 - val_loss: 0.1000 - val_accuracy: 0.9624 - 400ms/epoch - 11ms/step
Epoch 6/20
36/36 - 0s - loss: 0.0880 - accuracy: 0.9622 - val_loss: 0.0851 - val_accuracy: 0.9715 - 412ms/epoch - 11ms/step
Epoch 7/20
36/36 - 0s - loss: 0.0846 - accuracy: 0.9646 - val_loss: 0.0906 - val_accuracy: 0.9690 - 375ms/epoch - 10ms/step
Epoch 8/20
36/36 - 0s - loss: 0.0820 - accuracy: 0.9667 - val_loss: 0.0744 - val_accuracy: 0.9756 - 455ms/epoch - 13ms/step
Epoch 9/20
36/36 - 0s - loss: 0.0760 - accuracy: 0.9695 - val_loss: 0.0917 - val_accuracy: 0.9609 - 398ms/epoch - 11ms/step
Epoch 10/20
36/36 - 0s - loss: 0.0714 - accuracy: 0.9692 - val_loss: 0.0847 - val_accuracy: 0.9690 - 432ms/epoch - 12ms/step
Epoch 11/20
36/36 - 0s - loss: 0.0717 - accuracy: 0.9706 - val_loss: 0.0957 - val_accuracy: 0.9624 - 492ms/epoch - 14ms/step
Epoch 12/20
36/36 - 1s - loss: 0.0729 - accuracy: 0.9703 - val_loss: 0.0974 - val_accuracy: 0.9634 - 632ms/epoch - 18ms/step
Epoch 13/20
36/36 - 1s - loss: 0.0645 - accuracy: 0.9735 - val_loss: 0.0860 - val_accuracy: 0.9695 - 602ms/epoch - 17ms/step
Epoch 14/20
36/36 - 1s - loss: 0.0619 - accuracy: 0.9747 - val_loss: 0.0951 - val_accuracy: 0.9680 - 619ms/epoch - 17ms/step
Epoch 15/20
36/36 - 1s - loss: 0.0670 - accuracy: 0.9735 - val_loss: 0.0860 - val_accuracy: 0.9715 - 585ms/epoch - 16ms/step
Epoch 16/20
36/36 - 1s - loss: 0.0591 - accuracy: 0.9756 - val_loss: 0.0830 - val_accuracy: 0.9695 - 611ms/epoch - 17ms/step
Epoch 17/20
36/36 - 1s - loss: 0.0610 - accuracy: 0.9754 - val_loss: 0.0853 - val_accuracy: 0.9731 - 618ms/epoch - 17ms/step
Epoch 18/20
36/36 - 1s - loss: 0.0558 - accuracy: 0.9776 - val_loss: 0.0915 - val_accuracy: 0.9715 - 603ms/epoch - 17ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0556 - accuracy: 0.9795 - val_loss: 0.0877 - val_accuracy: 0.9710 - 496ms/epoch - 14ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0567 - accuracy: 0.9759 - val_loss: 0.0830 - val_accuracy: 0.9736 - 426ms/epoch - 12ms/step
Validation accuracy for Model of [Learning Rate 0.01 | Num Filters 64 | Dropout Rate 0.3 | FilterSize 3]: 0.9735772609710693
Epoch 1/20
36/36 - 1s - loss: 0.2399 - accuracy: 0.8983 - val_loss: 0.1445 - val_accuracy: 0.9507 - 1s/epoch - 37ms/step
Epoch 2/20
36/36 - 0s - loss: 0.1331 - accuracy: 0.9491 - val_loss: 0.1272 - val_accuracy: 0.9527 - 404ms/epoch - 11ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1109 - accuracy: 0.9562 - val_loss: 0.1028 - val_accuracy: 0.9660 - 413ms/epoch - 11ms/step
Epoch 4/20
36/36 - 0s - loss: 0.0962 - accuracy: 0.9603 - val_loss: 0.1112 - val_accuracy: 0.9558 - 400ms/epoch - 11ms/step
Epoch 5/20
36/36 - 0s - loss: 0.0952 - accuracy: 0.9614 - val_loss: 0.1122 - val_accuracy: 0.9517 - 399ms/epoch - 11ms/step
Epoch 6/20
36/36 - 0s - loss: 0.0826 - accuracy: 0.9655 - val_loss: 0.1112 - val_accuracy: 0.9573 - 423ms/epoch - 12ms/step
Epoch 7/20
36/36 - 0s - loss: 0.0846 - accuracy: 0.9655 - val_loss: 0.1061 - val_accuracy: 0.9583 - 361ms/epoch - 10ms/step
Epoch 8/20
36/36 - 0s - loss: 0.0727 - accuracy: 0.9697 - val_loss: 0.0889 - val_accuracy: 0.9644 - 401ms/epoch - 11ms/step
Epoch 9/20
36/36 - 0s - loss: 0.0760 - accuracy: 0.9688 - val_loss: 0.1095 - val_accuracy: 0.9548 - 418ms/epoch - 12ms/step
Epoch 10/20
36/36 - 0s - loss: 0.0685 - accuracy: 0.9730 - val_loss: 0.0959 - val_accuracy: 0.9593 - 411ms/epoch - 11ms/step
Epoch 11/20
36/36 - 0s - loss: 0.0675 - accuracy: 0.9734 - val_loss: 0.0949 - val_accuracy: 0.9675 - 408ms/epoch - 11ms/step
Epoch 12/20
36/36 - 0s - loss: 0.0693 - accuracy: 0.9731 - val_loss: 0.1059 - val_accuracy: 0.9568 - 405ms/epoch - 11ms/step
Epoch 13/20
36/36 - 0s - loss: 0.0654 - accuracy: 0.9719 - val_loss: 0.0926 - val_accuracy: 0.9649 - 377ms/epoch - 10ms/step
Epoch 14/20
36/36 - 0s - loss: 0.0586 - accuracy: 0.9759 - val_loss: 0.0929 - val_accuracy: 0.9700 - 409ms/epoch - 11ms/step
Epoch 15/20
36/36 - 0s - loss: 0.0556 - accuracy: 0.9780 - val_loss: 0.0989 - val_accuracy: 0.9721 - 372ms/epoch - 10ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0571 - accuracy: 0.9779 - val_loss: 0.0854 - val_accuracy: 0.9731 - 416ms/epoch - 12ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0551 - accuracy: 0.9795 - val_loss: 0.1098 - val_accuracy: 0.9604 - 370ms/epoch - 10ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0599 - accuracy: 0.9764 - val_loss: 0.0898 - val_accuracy: 0.9721 - 368ms/epoch - 10ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0552 - accuracy: 0.9774 - val_loss: 0.0812 - val_accuracy: 0.9776 - 366ms/epoch - 10ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0531 - accuracy: 0.9785 - val_loss: 0.1004 - val_accuracy: 0.9644 - 397ms/epoch - 11ms/step
Validation accuracy for Model of [Learning Rate 0.01 | Num Filters 64 | Dropout Rate 0.3 | FilterSize 5]: 0.9644308686256409
Epoch 1/20
36/36 - 2s - loss: 0.2289 - accuracy: 0.9029 - val_loss: 0.1176 - val_accuracy: 0.9563 - 2s/epoch - 44ms/step
Epoch 2/20
36/36 - 0s - loss: 0.1315 - accuracy: 0.9476 - val_loss: 0.1190 - val_accuracy: 0.9548 - 353ms/epoch - 10ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1086 - accuracy: 0.9554 - val_loss: 0.1369 - val_accuracy: 0.9482 - 389ms/epoch - 11ms/step
Epoch 4/20
36/36 - 0s - loss: 0.0913 - accuracy: 0.9632 - val_loss: 0.0902 - val_accuracy: 0.9665 - 416ms/epoch - 12ms/step
Epoch 5/20
36/36 - 0s - loss: 0.0871 - accuracy: 0.9647 - val_loss: 0.1104 - val_accuracy: 0.9604 - 351ms/epoch - 10ms/step
Epoch 6/20
36/36 - 0s - loss: 0.0788 - accuracy: 0.9661 - val_loss: 0.0975 - val_accuracy: 0.9614 - 377ms/epoch - 10ms/step
Epoch 7/20
36/36 - 0s - loss: 0.0744 - accuracy: 0.9694 - val_loss: 0.0974 - val_accuracy: 0.9639 - 402ms/epoch - 11ms/step
Epoch 8/20
36/36 - 0s - loss: 0.0779 - accuracy: 0.9693 - val_loss: 0.0836 - val_accuracy: 0.9721 - 338ms/epoch - 9ms/step
Epoch 9/20
36/36 - 0s - loss: 0.0728 - accuracy: 0.9710 - val_loss: 0.0874 - val_accuracy: 0.9695 - 411ms/epoch - 11ms/step
Epoch 10/20
36/36 - 0s - loss: 0.0651 - accuracy: 0.9734 - val_loss: 0.0805 - val_accuracy: 0.9766 - 390ms/epoch - 11ms/step
Epoch 11/20
36/36 - 0s - loss: 0.0638 - accuracy: 0.9750 - val_loss: 0.0944 - val_accuracy: 0.9715 - 405ms/epoch - 11ms/step
Epoch 12/20
36/36 - 0s - loss: 0.0625 - accuracy: 0.9743 - val_loss: 0.0873 - val_accuracy: 0.9746 - 400ms/epoch - 11ms/step
Epoch 13/20
36/36 - 0s - loss: 0.0621 - accuracy: 0.9759 - val_loss: 0.0848 - val_accuracy: 0.9731 - 345ms/epoch - 10ms/step
Epoch 14/20
36/36 - 0s - loss: 0.0551 - accuracy: 0.9781 - val_loss: 0.1147 - val_accuracy: 0.9563 - 398ms/epoch - 11ms/step
Epoch 15/20
36/36 - 0s - loss: 0.0621 - accuracy: 0.9759 - val_loss: 0.0878 - val_accuracy: 0.9776 - 351ms/epoch - 10ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0607 - accuracy: 0.9763 - val_loss: 0.0899 - val_accuracy: 0.9700 - 381ms/epoch - 11ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0537 - accuracy: 0.9796 - val_loss: 0.0850 - val_accuracy: 0.9736 - 408ms/epoch - 11ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0489 - accuracy: 0.9823 - val_loss: 0.0908 - val_accuracy: 0.9695 - 356ms/epoch - 10ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0464 - accuracy: 0.9828 - val_loss: 0.1069 - val_accuracy: 0.9695 - 403ms/epoch - 11ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0480 - accuracy: 0.9815 - val_loss: 0.0972 - val_accuracy: 0.9639 - 406ms/epoch - 11ms/step
Validation accuracy for Model of [Learning Rate 0.01 | Num Filters 64 | Dropout Rate 0.3 | FilterSize 7]: 0.9639227390289307
Epoch 1/20
36/36 - 2s - loss: 0.2386 - accuracy: 0.9030 - val_loss: 0.1614 - val_accuracy: 0.9436 - 2s/epoch - 56ms/step
Epoch 2/20
36/36 - 1s - loss: 0.1429 - accuracy: 0.9473 - val_loss: 0.1444 - val_accuracy: 0.9421 - 627ms/epoch - 17ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1279 - accuracy: 0.9512 - val_loss: 0.1298 - val_accuracy: 0.9472 - 647ms/epoch - 18ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1208 - accuracy: 0.9522 - val_loss: 0.1332 - val_accuracy: 0.9365 - 521ms/epoch - 14ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1043 - accuracy: 0.9574 - val_loss: 0.1034 - val_accuracy: 0.9588 - 378ms/epoch - 11ms/step
Epoch 6/20
36/36 - 0s - loss: 0.0982 - accuracy: 0.9613 - val_loss: 0.0956 - val_accuracy: 0.9609 - 389ms/epoch - 11ms/step
Epoch 7/20
36/36 - 0s - loss: 0.0951 - accuracy: 0.9624 - val_loss: 0.0951 - val_accuracy: 0.9675 - 412ms/epoch - 11ms/step
Epoch 8/20
36/36 - 0s - loss: 0.0854 - accuracy: 0.9640 - val_loss: 0.0864 - val_accuracy: 0.9721 - 426ms/epoch - 12ms/step
Epoch 9/20
36/36 - 0s - loss: 0.0885 - accuracy: 0.9654 - val_loss: 0.0845 - val_accuracy: 0.9715 - 406ms/epoch - 11ms/step
Epoch 10/20
36/36 - 0s - loss: 0.0851 - accuracy: 0.9648 - val_loss: 0.0932 - val_accuracy: 0.9665 - 408ms/epoch - 11ms/step
Epoch 11/20
36/36 - 0s - loss: 0.0777 - accuracy: 0.9678 - val_loss: 0.0898 - val_accuracy: 0.9654 - 397ms/epoch - 11ms/step
Epoch 12/20
36/36 - 0s - loss: 0.0790 - accuracy: 0.9680 - val_loss: 0.0924 - val_accuracy: 0.9721 - 373ms/epoch - 10ms/step
Epoch 13/20
36/36 - 0s - loss: 0.0817 - accuracy: 0.9672 - val_loss: 0.0994 - val_accuracy: 0.9578 - 401ms/epoch - 11ms/step
Epoch 14/20
36/36 - 0s - loss: 0.0743 - accuracy: 0.9698 - val_loss: 0.0809 - val_accuracy: 0.9705 - 380ms/epoch - 11ms/step
Epoch 15/20
36/36 - 0s - loss: 0.0817 - accuracy: 0.9657 - val_loss: 0.0870 - val_accuracy: 0.9726 - 387ms/epoch - 11ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0785 - accuracy: 0.9688 - val_loss: 0.0900 - val_accuracy: 0.9690 - 394ms/epoch - 11ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0729 - accuracy: 0.9704 - val_loss: 0.0879 - val_accuracy: 0.9685 - 371ms/epoch - 10ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0770 - accuracy: 0.9673 - val_loss: 0.0925 - val_accuracy: 0.9685 - 421ms/epoch - 12ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0738 - accuracy: 0.9690 - val_loss: 0.1002 - val_accuracy: 0.9649 - 372ms/epoch - 10ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0714 - accuracy: 0.9720 - val_loss: 0.1065 - val_accuracy: 0.9578 - 376ms/epoch - 10ms/step
Validation accuracy for Model of [Learning Rate 0.01 | Num Filters 64 | Dropout Rate 0.5 | FilterSize 3]: 0.9578251838684082
Epoch 1/20
36/36 - 1s - loss: 0.2399 - accuracy: 0.9021 - val_loss: 0.1286 - val_accuracy: 0.9548 - 1s/epoch - 38ms/step
Epoch 2/20
36/36 - 0s - loss: 0.1401 - accuracy: 0.9453 - val_loss: 0.1066 - val_accuracy: 0.9619 - 414ms/epoch - 12ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1164 - accuracy: 0.9539 - val_loss: 0.1253 - val_accuracy: 0.9538 - 374ms/epoch - 10ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1095 - accuracy: 0.9563 - val_loss: 0.1182 - val_accuracy: 0.9538 - 552ms/epoch - 15ms/step
Epoch 5/20
36/36 - 1s - loss: 0.0968 - accuracy: 0.9621 - val_loss: 0.1022 - val_accuracy: 0.9588 - 616ms/epoch - 17ms/step
Epoch 6/20
36/36 - 1s - loss: 0.0879 - accuracy: 0.9637 - val_loss: 0.1157 - val_accuracy: 0.9553 - 573ms/epoch - 16ms/step
Epoch 7/20
36/36 - 1s - loss: 0.0877 - accuracy: 0.9648 - val_loss: 0.0894 - val_accuracy: 0.9710 - 626ms/epoch - 17ms/step
Epoch 8/20
36/36 - 1s - loss: 0.0857 - accuracy: 0.9658 - val_loss: 0.1186 - val_accuracy: 0.9482 - 608ms/epoch - 17ms/step
Epoch 9/20
36/36 - 1s - loss: 0.0818 - accuracy: 0.9656 - val_loss: 0.0913 - val_accuracy: 0.9705 - 602ms/epoch - 17ms/step
Epoch 10/20
36/36 - 1s - loss: 0.0783 - accuracy: 0.9682 - val_loss: 0.0877 - val_accuracy: 0.9685 - 639ms/epoch - 18ms/step
Epoch 11/20
36/36 - 1s - loss: 0.0781 - accuracy: 0.9691 - val_loss: 0.1196 - val_accuracy: 0.9527 - 558ms/epoch - 15ms/step
Epoch 12/20
36/36 - 0s - loss: 0.0781 - accuracy: 0.9690 - val_loss: 0.1123 - val_accuracy: 0.9588 - 404ms/epoch - 11ms/step
Epoch 13/20
36/36 - 0s - loss: 0.0810 - accuracy: 0.9679 - val_loss: 0.0897 - val_accuracy: 0.9705 - 368ms/epoch - 10ms/step
Epoch 14/20
36/36 - 0s - loss: 0.0720 - accuracy: 0.9689 - val_loss: 0.1019 - val_accuracy: 0.9614 - 370ms/epoch - 10ms/step
Epoch 15/20
36/36 - 0s - loss: 0.0752 - accuracy: 0.9699 - val_loss: 0.0954 - val_accuracy: 0.9670 - 378ms/epoch - 10ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0763 - accuracy: 0.9693 - val_loss: 0.0907 - val_accuracy: 0.9665 - 404ms/epoch - 11ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0753 - accuracy: 0.9708 - val_loss: 0.0951 - val_accuracy: 0.9629 - 363ms/epoch - 10ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0697 - accuracy: 0.9724 - val_loss: 0.0955 - val_accuracy: 0.9675 - 397ms/epoch - 11ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0707 - accuracy: 0.9729 - val_loss: 0.0956 - val_accuracy: 0.9741 - 364ms/epoch - 10ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0713 - accuracy: 0.9723 - val_loss: 0.0901 - val_accuracy: 0.9680 - 401ms/epoch - 11ms/step
Validation accuracy for Model of [Learning Rate 0.01 | Num Filters 64 | Dropout Rate 0.5 | FilterSize 5]: 0.9679877758026123
Epoch 1/20
36/36 - 1s - loss: 0.2710 - accuracy: 0.8802 - val_loss: 0.1933 - val_accuracy: 0.9278 - 1s/epoch - 38ms/step
Epoch 2/20
36/36 - 0s - loss: 0.1453 - accuracy: 0.9436 - val_loss: 0.1048 - val_accuracy: 0.9593 - 402ms/epoch - 11ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1149 - accuracy: 0.9561 - val_loss: 0.0999 - val_accuracy: 0.9654 - 337ms/epoch - 9ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1091 - accuracy: 0.9563 - val_loss: 0.1052 - val_accuracy: 0.9609 - 362ms/epoch - 10ms/step
Epoch 5/20
36/36 - 0s - loss: 0.0990 - accuracy: 0.9605 - val_loss: 0.1407 - val_accuracy: 0.9482 - 342ms/epoch - 9ms/step
Epoch 6/20
36/36 - 0s - loss: 0.0876 - accuracy: 0.9645 - val_loss: 0.0935 - val_accuracy: 0.9639 - 338ms/epoch - 9ms/step
Epoch 7/20
36/36 - 0s - loss: 0.0861 - accuracy: 0.9641 - val_loss: 0.0926 - val_accuracy: 0.9690 - 358ms/epoch - 10ms/step
Epoch 8/20
36/36 - 0s - loss: 0.0790 - accuracy: 0.9677 - val_loss: 0.1067 - val_accuracy: 0.9593 - 389ms/epoch - 11ms/step
Epoch 9/20
36/36 - 0s - loss: 0.0837 - accuracy: 0.9663 - val_loss: 0.1262 - val_accuracy: 0.9441 - 397ms/epoch - 11ms/step
Epoch 10/20
36/36 - 0s - loss: 0.0793 - accuracy: 0.9687 - val_loss: 0.0873 - val_accuracy: 0.9710 - 395ms/epoch - 11ms/step
Epoch 11/20
36/36 - 0s - loss: 0.0744 - accuracy: 0.9700 - val_loss: 0.0881 - val_accuracy: 0.9690 - 386ms/epoch - 11ms/step
Epoch 12/20
36/36 - 0s - loss: 0.0708 - accuracy: 0.9716 - val_loss: 0.1126 - val_accuracy: 0.9502 - 369ms/epoch - 10ms/step
Epoch 13/20
36/36 - 0s - loss: 0.0697 - accuracy: 0.9711 - val_loss: 0.1000 - val_accuracy: 0.9649 - 451ms/epoch - 13ms/step
Epoch 14/20
36/36 - 1s - loss: 0.0677 - accuracy: 0.9729 - val_loss: 0.0994 - val_accuracy: 0.9660 - 582ms/epoch - 16ms/step
Epoch 15/20
36/36 - 1s - loss: 0.0646 - accuracy: 0.9726 - val_loss: 0.1017 - val_accuracy: 0.9670 - 590ms/epoch - 16ms/step
Epoch 16/20
36/36 - 1s - loss: 0.0668 - accuracy: 0.9730 - val_loss: 0.0978 - val_accuracy: 0.9690 - 585ms/epoch - 16ms/step
Epoch 17/20
36/36 - 1s - loss: 0.0647 - accuracy: 0.9737 - val_loss: 0.0932 - val_accuracy: 0.9710 - 551ms/epoch - 15ms/step
Epoch 18/20
36/36 - 1s - loss: 0.0625 - accuracy: 0.9746 - val_loss: 0.0874 - val_accuracy: 0.9746 - 589ms/epoch - 16ms/step
Epoch 19/20
36/36 - 1s - loss: 0.0646 - accuracy: 0.9738 - val_loss: 0.0943 - val_accuracy: 0.9736 - 545ms/epoch - 15ms/step
Epoch 20/20
36/36 - 1s - loss: 0.0647 - accuracy: 0.9758 - val_loss: 0.0890 - val_accuracy: 0.9726 - 567ms/epoch - 16ms/step
Validation accuracy for Model of [Learning Rate 0.01 | Num Filters 64 | Dropout Rate 0.5 | FilterSize 7]: 0.9725610017776489
Epoch 1/20
36/36 - 2s - loss: 0.2612 - accuracy: 0.8930 - val_loss: 0.1490 - val_accuracy: 0.9517 - 2s/epoch - 68ms/step
Epoch 2/20
36/36 - 0s - loss: 0.1666 - accuracy: 0.9354 - val_loss: 0.1922 - val_accuracy: 0.9177 - 394ms/epoch - 11ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1498 - accuracy: 0.9432 - val_loss: 0.1357 - val_accuracy: 0.9441 - 374ms/epoch - 10ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1304 - accuracy: 0.9484 - val_loss: 0.1155 - val_accuracy: 0.9558 - 448ms/epoch - 12ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1297 - accuracy: 0.9471 - val_loss: 0.1521 - val_accuracy: 0.9421 - 420ms/epoch - 12ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1296 - accuracy: 0.9503 - val_loss: 0.1245 - val_accuracy: 0.9472 - 439ms/epoch - 12ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1213 - accuracy: 0.9512 - val_loss: 0.1171 - val_accuracy: 0.9522 - 411ms/epoch - 11ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1206 - accuracy: 0.9503 - val_loss: 0.1119 - val_accuracy: 0.9553 - 394ms/epoch - 11ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1094 - accuracy: 0.9560 - val_loss: 0.1349 - val_accuracy: 0.9405 - 436ms/epoch - 12ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1120 - accuracy: 0.9552 - val_loss: 0.1098 - val_accuracy: 0.9568 - 384ms/epoch - 11ms/step
Epoch 11/20
36/36 - 0s - loss: 0.1046 - accuracy: 0.9566 - val_loss: 0.0990 - val_accuracy: 0.9634 - 446ms/epoch - 12ms/step
Epoch 12/20
36/36 - 0s - loss: 0.1027 - accuracy: 0.9559 - val_loss: 0.0936 - val_accuracy: 0.9680 - 419ms/epoch - 12ms/step
Epoch 13/20
36/36 - 0s - loss: 0.1052 - accuracy: 0.9551 - val_loss: 0.1104 - val_accuracy: 0.9558 - 398ms/epoch - 11ms/step
Epoch 14/20
36/36 - 0s - loss: 0.1058 - accuracy: 0.9546 - val_loss: 0.1096 - val_accuracy: 0.9629 - 420ms/epoch - 12ms/step
Epoch 15/20
36/36 - 0s - loss: 0.1051 - accuracy: 0.9547 - val_loss: 0.1080 - val_accuracy: 0.9543 - 429ms/epoch - 12ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0976 - accuracy: 0.9595 - val_loss: 0.1027 - val_accuracy: 0.9573 - 438ms/epoch - 12ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0993 - accuracy: 0.9594 - val_loss: 0.1005 - val_accuracy: 0.9665 - 451ms/epoch - 13ms/step
Epoch 18/20
36/36 - 1s - loss: 0.0938 - accuracy: 0.9633 - val_loss: 0.0962 - val_accuracy: 0.9710 - 625ms/epoch - 17ms/step
Epoch 19/20
36/36 - 1s - loss: 0.0963 - accuracy: 0.9622 - val_loss: 0.0980 - val_accuracy: 0.9670 - 608ms/epoch - 17ms/step
Epoch 20/20
36/36 - 1s - loss: 0.0968 - accuracy: 0.9589 - val_loss: 0.1029 - val_accuracy: 0.9649 - 619ms/epoch - 17ms/step
Validation accuracy for Model of [Learning Rate 0.01 | Num Filters 64 | Dropout Rate 0.7 | FilterSize 3]: 0.9649389982223511
Epoch 1/20
36/36 - 2s - loss: 0.2853 - accuracy: 0.8843 - val_loss: 0.1400 - val_accuracy: 0.9502 - 2s/epoch - 58ms/step
Epoch 2/20
36/36 - 1s - loss: 0.1570 - accuracy: 0.9401 - val_loss: 0.1425 - val_accuracy: 0.9492 - 500ms/epoch - 14ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1364 - accuracy: 0.9486 - val_loss: 0.1073 - val_accuracy: 0.9593 - 380ms/epoch - 11ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1230 - accuracy: 0.9514 - val_loss: 0.1119 - val_accuracy: 0.9527 - 403ms/epoch - 11ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1181 - accuracy: 0.9546 - val_loss: 0.1254 - val_accuracy: 0.9533 - 414ms/epoch - 11ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1179 - accuracy: 0.9525 - val_loss: 0.1272 - val_accuracy: 0.9497 - 422ms/epoch - 12ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1156 - accuracy: 0.9546 - val_loss: 0.1013 - val_accuracy: 0.9619 - 394ms/epoch - 11ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1065 - accuracy: 0.9564 - val_loss: 0.1108 - val_accuracy: 0.9522 - 393ms/epoch - 11ms/step
Epoch 9/20
36/36 - 1s - loss: 0.1058 - accuracy: 0.9577 - val_loss: 0.0965 - val_accuracy: 0.9665 - 506ms/epoch - 14ms/step
Epoch 10/20
36/36 - 1s - loss: 0.1063 - accuracy: 0.9581 - val_loss: 0.0994 - val_accuracy: 0.9593 - 634ms/epoch - 18ms/step
Epoch 11/20
36/36 - 1s - loss: 0.0986 - accuracy: 0.9602 - val_loss: 0.1102 - val_accuracy: 0.9543 - 599ms/epoch - 17ms/step
Epoch 12/20
36/36 - 1s - loss: 0.0967 - accuracy: 0.9619 - val_loss: 0.1079 - val_accuracy: 0.9634 - 592ms/epoch - 16ms/step
Epoch 13/20
36/36 - 1s - loss: 0.0987 - accuracy: 0.9602 - val_loss: 0.1104 - val_accuracy: 0.9548 - 591ms/epoch - 16ms/step
Epoch 14/20
36/36 - 1s - loss: 0.0950 - accuracy: 0.9636 - val_loss: 0.1222 - val_accuracy: 0.9466 - 585ms/epoch - 16ms/step
Epoch 15/20
36/36 - 1s - loss: 0.0946 - accuracy: 0.9620 - val_loss: 0.1014 - val_accuracy: 0.9670 - 616ms/epoch - 17ms/step
Epoch 16/20
36/36 - 1s - loss: 0.0894 - accuracy: 0.9635 - val_loss: 0.1115 - val_accuracy: 0.9583 - 629ms/epoch - 17ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0923 - accuracy: 0.9621 - val_loss: 0.1288 - val_accuracy: 0.9517 - 454ms/epoch - 13ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0928 - accuracy: 0.9621 - val_loss: 0.1066 - val_accuracy: 0.9624 - 390ms/epoch - 11ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0931 - accuracy: 0.9619 - val_loss: 0.0994 - val_accuracy: 0.9675 - 427ms/epoch - 12ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0926 - accuracy: 0.9626 - val_loss: 0.1030 - val_accuracy: 0.9558 - 407ms/epoch - 11ms/step
Validation accuracy for Model of [Learning Rate 0.01 | Num Filters 64 | Dropout Rate 0.7 | FilterSize 5]: 0.9557926654815674
Epoch 1/20
36/36 - 1s - loss: 0.2827 - accuracy: 0.8786 - val_loss: 0.1399 - val_accuracy: 0.9461 - 1s/epoch - 38ms/step
Epoch 2/20
36/36 - 0s - loss: 0.1556 - accuracy: 0.9414 - val_loss: 0.1252 - val_accuracy: 0.9548 - 365ms/epoch - 10ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1317 - accuracy: 0.9508 - val_loss: 0.1115 - val_accuracy: 0.9533 - 407ms/epoch - 11ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1176 - accuracy: 0.9534 - val_loss: 0.1341 - val_accuracy: 0.9461 - 371ms/epoch - 10ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1101 - accuracy: 0.9559 - val_loss: 0.0965 - val_accuracy: 0.9654 - 348ms/epoch - 10ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1120 - accuracy: 0.9549 - val_loss: 0.1095 - val_accuracy: 0.9533 - 351ms/epoch - 10ms/step
Epoch 7/20
36/36 - 1s - loss: 0.1113 - accuracy: 0.9551 - val_loss: 0.1225 - val_accuracy: 0.9502 - 550ms/epoch - 15ms/step
Epoch 8/20
36/36 - 1s - loss: 0.0977 - accuracy: 0.9574 - val_loss: 0.1139 - val_accuracy: 0.9573 - 569ms/epoch - 16ms/step
Epoch 9/20
36/36 - 1s - loss: 0.0951 - accuracy: 0.9622 - val_loss: 0.1134 - val_accuracy: 0.9558 - 562ms/epoch - 16ms/step
Epoch 10/20
36/36 - 1s - loss: 0.0934 - accuracy: 0.9610 - val_loss: 0.1025 - val_accuracy: 0.9639 - 576ms/epoch - 16ms/step
Epoch 11/20
36/36 - 1s - loss: 0.0947 - accuracy: 0.9606 - val_loss: 0.0921 - val_accuracy: 0.9665 - 556ms/epoch - 15ms/step
Epoch 12/20
36/36 - 1s - loss: 0.0902 - accuracy: 0.9623 - val_loss: 0.0964 - val_accuracy: 0.9609 - 594ms/epoch - 17ms/step
Epoch 13/20
36/36 - 1s - loss: 0.0972 - accuracy: 0.9615 - val_loss: 0.1172 - val_accuracy: 0.9527 - 582ms/epoch - 16ms/step
Epoch 14/20
36/36 - 1s - loss: 0.0926 - accuracy: 0.9610 - val_loss: 0.0973 - val_accuracy: 0.9654 - 590ms/epoch - 16ms/step
Epoch 15/20
36/36 - 0s - loss: 0.0879 - accuracy: 0.9633 - val_loss: 0.1032 - val_accuracy: 0.9634 - 454ms/epoch - 13ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0872 - accuracy: 0.9621 - val_loss: 0.1316 - val_accuracy: 0.9472 - 399ms/epoch - 11ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0864 - accuracy: 0.9634 - val_loss: 0.0991 - val_accuracy: 0.9649 - 379ms/epoch - 11ms/step
Epoch 18/20
36/36 - 0s - loss: 0.0865 - accuracy: 0.9629 - val_loss: 0.1035 - val_accuracy: 0.9665 - 358ms/epoch - 10ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0785 - accuracy: 0.9673 - val_loss: 0.1032 - val_accuracy: 0.9604 - 405ms/epoch - 11ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0766 - accuracy: 0.9689 - val_loss: 0.0937 - val_accuracy: 0.9726 - 364ms/epoch - 10ms/step
Validation accuracy for Model of [Learning Rate 0.01 | Num Filters 64 | Dropout Rate 0.7 | FilterSize 7]: 0.9725610017776489
Epoch 1/20
36/36 - 2s - loss: 0.2372 - accuracy: 0.9027 - val_loss: 0.1287 - val_accuracy: 0.9538 - 2s/epoch - 43ms/step
Epoch 2/20
36/36 - 1s - loss: 0.1331 - accuracy: 0.9494 - val_loss: 0.1159 - val_accuracy: 0.9538 - 604ms/epoch - 17ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1103 - accuracy: 0.9551 - val_loss: 0.1057 - val_accuracy: 0.9573 - 607ms/epoch - 17ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1055 - accuracy: 0.9566 - val_loss: 0.0922 - val_accuracy: 0.9639 - 599ms/epoch - 17ms/step
Epoch 5/20
36/36 - 1s - loss: 0.1023 - accuracy: 0.9599 - val_loss: 0.1173 - val_accuracy: 0.9456 - 602ms/epoch - 17ms/step
Epoch 6/20
36/36 - 1s - loss: 0.0895 - accuracy: 0.9637 - val_loss: 0.0956 - val_accuracy: 0.9670 - 601ms/epoch - 17ms/step
Epoch 7/20
36/36 - 1s - loss: 0.0858 - accuracy: 0.9657 - val_loss: 0.1122 - val_accuracy: 0.9477 - 607ms/epoch - 17ms/step
Epoch 8/20
36/36 - 1s - loss: 0.0780 - accuracy: 0.9683 - val_loss: 0.0952 - val_accuracy: 0.9634 - 610ms/epoch - 17ms/step
Epoch 9/20
36/36 - 1s - loss: 0.0759 - accuracy: 0.9697 - val_loss: 0.0979 - val_accuracy: 0.9654 - 629ms/epoch - 17ms/step
Epoch 10/20
36/36 - 1s - loss: 0.0745 - accuracy: 0.9704 - val_loss: 0.0784 - val_accuracy: 0.9710 - 598ms/epoch - 17ms/step
Epoch 11/20
36/36 - 1s - loss: 0.0757 - accuracy: 0.9689 - val_loss: 0.0792 - val_accuracy: 0.9721 - 629ms/epoch - 17ms/step
Epoch 12/20
36/36 - 1s - loss: 0.0689 - accuracy: 0.9735 - val_loss: 0.0862 - val_accuracy: 0.9715 - 976ms/epoch - 27ms/step
Epoch 13/20
36/36 - 1s - loss: 0.0704 - accuracy: 0.9731 - val_loss: 0.0886 - val_accuracy: 0.9690 - 986ms/epoch - 27ms/step
Epoch 14/20
36/36 - 1s - loss: 0.0662 - accuracy: 0.9741 - val_loss: 0.0896 - val_accuracy: 0.9726 - 974ms/epoch - 27ms/step
Epoch 15/20
36/36 - 1s - loss: 0.0606 - accuracy: 0.9758 - val_loss: 0.0823 - val_accuracy: 0.9715 - 1s/epoch - 29ms/step
Epoch 16/20
36/36 - 1s - loss: 0.0609 - accuracy: 0.9759 - val_loss: 0.0819 - val_accuracy: 0.9756 - 894ms/epoch - 25ms/step
Epoch 17/20
36/36 - 1s - loss: 0.0630 - accuracy: 0.9754 - val_loss: 0.0778 - val_accuracy: 0.9766 - 614ms/epoch - 17ms/step
Epoch 18/20
36/36 - 1s - loss: 0.0590 - accuracy: 0.9766 - val_loss: 0.1119 - val_accuracy: 0.9644 - 610ms/epoch - 17ms/step
Epoch 19/20
36/36 - 1s - loss: 0.0579 - accuracy: 0.9770 - val_loss: 0.0893 - val_accuracy: 0.9695 - 585ms/epoch - 16ms/step
Epoch 20/20
36/36 - 1s - loss: 0.0666 - accuracy: 0.9765 - val_loss: 0.1063 - val_accuracy: 0.9670 - 603ms/epoch - 17ms/step
Validation accuracy for Model of [Learning Rate 0.01 | Num Filters 128 | Dropout Rate 0.3 | FilterSize 3]: 0.9669715166091919
Epoch 1/20
36/36 - 2s - loss: 0.2426 - accuracy: 0.9019 - val_loss: 0.1421 - val_accuracy: 0.9482 - 2s/epoch - 52ms/step
Epoch 2/20
36/36 - 1s - loss: 0.1333 - accuracy: 0.9467 - val_loss: 0.1080 - val_accuracy: 0.9593 - 932ms/epoch - 26ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1075 - accuracy: 0.9557 - val_loss: 0.0977 - val_accuracy: 0.9593 - 925ms/epoch - 26ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1004 - accuracy: 0.9580 - val_loss: 0.0959 - val_accuracy: 0.9680 - 949ms/epoch - 26ms/step
Epoch 5/20
36/36 - 1s - loss: 0.0939 - accuracy: 0.9610 - val_loss: 0.0873 - val_accuracy: 0.9680 - 945ms/epoch - 26ms/step
Epoch 6/20
36/36 - 1s - loss: 0.0871 - accuracy: 0.9660 - val_loss: 0.1071 - val_accuracy: 0.9614 - 651ms/epoch - 18ms/step
Epoch 7/20
36/36 - 1s - loss: 0.0839 - accuracy: 0.9669 - val_loss: 0.0936 - val_accuracy: 0.9619 - 610ms/epoch - 17ms/step
Epoch 8/20
36/36 - 1s - loss: 0.0835 - accuracy: 0.9679 - val_loss: 0.0868 - val_accuracy: 0.9700 - 563ms/epoch - 16ms/step
Epoch 9/20
36/36 - 1s - loss: 0.0782 - accuracy: 0.9692 - val_loss: 0.0895 - val_accuracy: 0.9685 - 572ms/epoch - 16ms/step
Epoch 10/20
36/36 - 1s - loss: 0.0665 - accuracy: 0.9748 - val_loss: 0.0923 - val_accuracy: 0.9695 - 558ms/epoch - 16ms/step
Epoch 11/20
36/36 - 1s - loss: 0.0701 - accuracy: 0.9708 - val_loss: 0.0953 - val_accuracy: 0.9614 - 601ms/epoch - 17ms/step
Epoch 12/20
36/36 - 1s - loss: 0.0635 - accuracy: 0.9757 - val_loss: 0.0938 - val_accuracy: 0.9726 - 571ms/epoch - 16ms/step
Epoch 13/20
36/36 - 1s - loss: 0.0746 - accuracy: 0.9712 - val_loss: 0.1082 - val_accuracy: 0.9553 - 565ms/epoch - 16ms/step
Epoch 14/20
36/36 - 1s - loss: 0.0711 - accuracy: 0.9716 - val_loss: 0.0996 - val_accuracy: 0.9614 - 607ms/epoch - 17ms/step
Epoch 15/20
36/36 - 1s - loss: 0.0596 - accuracy: 0.9756 - val_loss: 0.0882 - val_accuracy: 0.9736 - 604ms/epoch - 17ms/step
Epoch 16/20
36/36 - 1s - loss: 0.0593 - accuracy: 0.9755 - val_loss: 0.0975 - val_accuracy: 0.9588 - 578ms/epoch - 16ms/step
Epoch 17/20
36/36 - 1s - loss: 0.0553 - accuracy: 0.9778 - val_loss: 0.0889 - val_accuracy: 0.9695 - 596ms/epoch - 17ms/step
Epoch 18/20
36/36 - 1s - loss: 0.0573 - accuracy: 0.9748 - val_loss: 0.1097 - val_accuracy: 0.9634 - 584ms/epoch - 16ms/step
Epoch 19/20
36/36 - 1s - loss: 0.0539 - accuracy: 0.9788 - val_loss: 0.0893 - val_accuracy: 0.9690 - 620ms/epoch - 17ms/step
Epoch 20/20
36/36 - 1s - loss: 0.0495 - accuracy: 0.9803 - val_loss: 0.0943 - val_accuracy: 0.9746 - 612ms/epoch - 17ms/step
Validation accuracy for Model of [Learning Rate 0.01 | Num Filters 128 | Dropout Rate 0.3 | FilterSize 5]: 0.9745935201644897
Epoch 1/20
36/36 - 2s - loss: 0.2557 - accuracy: 0.8880 - val_loss: 0.1619 - val_accuracy: 0.9416 - 2s/epoch - 48ms/step
Epoch 2/20
36/36 - 1s - loss: 0.1217 - accuracy: 0.9522 - val_loss: 0.1245 - val_accuracy: 0.9548 - 864ms/epoch - 24ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1014 - accuracy: 0.9590 - val_loss: 0.1045 - val_accuracy: 0.9578 - 874ms/epoch - 24ms/step
Epoch 4/20
36/36 - 1s - loss: 0.0915 - accuracy: 0.9652 - val_loss: 0.0874 - val_accuracy: 0.9690 - 840ms/epoch - 23ms/step
Epoch 5/20
36/36 - 1s - loss: 0.0881 - accuracy: 0.9634 - val_loss: 0.1068 - val_accuracy: 0.9604 - 853ms/epoch - 24ms/step
Epoch 6/20
36/36 - 1s - loss: 0.0810 - accuracy: 0.9671 - val_loss: 0.0850 - val_accuracy: 0.9690 - 774ms/epoch - 21ms/step
Epoch 7/20
36/36 - 1s - loss: 0.0768 - accuracy: 0.9676 - val_loss: 0.1045 - val_accuracy: 0.9660 - 541ms/epoch - 15ms/step
Epoch 8/20
36/36 - 1s - loss: 0.0706 - accuracy: 0.9718 - val_loss: 0.0842 - val_accuracy: 0.9660 - 550ms/epoch - 15ms/step
Epoch 9/20
36/36 - 1s - loss: 0.0740 - accuracy: 0.9705 - val_loss: 0.1141 - val_accuracy: 0.9568 - 551ms/epoch - 15ms/step
Epoch 10/20
36/36 - 1s - loss: 0.0688 - accuracy: 0.9733 - val_loss: 0.0906 - val_accuracy: 0.9665 - 524ms/epoch - 15ms/step
Epoch 11/20
36/36 - 1s - loss: 0.0611 - accuracy: 0.9755 - val_loss: 0.0956 - val_accuracy: 0.9629 - 571ms/epoch - 16ms/step
Epoch 12/20
36/36 - 1s - loss: 0.0614 - accuracy: 0.9777 - val_loss: 0.0969 - val_accuracy: 0.9665 - 548ms/epoch - 15ms/step
Epoch 13/20
36/36 - 1s - loss: 0.0590 - accuracy: 0.9763 - val_loss: 0.0905 - val_accuracy: 0.9690 - 527ms/epoch - 15ms/step
Epoch 14/20
36/36 - 1s - loss: 0.0627 - accuracy: 0.9746 - val_loss: 0.0921 - val_accuracy: 0.9685 - 547ms/epoch - 15ms/step
Epoch 15/20
36/36 - 1s - loss: 0.0632 - accuracy: 0.9767 - val_loss: 0.0905 - val_accuracy: 0.9700 - 534ms/epoch - 15ms/step
Epoch 16/20
36/36 - 1s - loss: 0.0562 - accuracy: 0.9781 - val_loss: 0.1018 - val_accuracy: 0.9644 - 521ms/epoch - 14ms/step
Epoch 17/20
36/36 - 1s - loss: 0.0529 - accuracy: 0.9793 - val_loss: 0.1020 - val_accuracy: 0.9695 - 571ms/epoch - 16ms/step
Epoch 18/20
36/36 - 1s - loss: 0.0529 - accuracy: 0.9799 - val_loss: 0.1021 - val_accuracy: 0.9649 - 545ms/epoch - 15ms/step
Epoch 19/20
36/36 - 1s - loss: 0.0496 - accuracy: 0.9802 - val_loss: 0.0823 - val_accuracy: 0.9761 - 560ms/epoch - 16ms/step
Epoch 20/20
36/36 - 1s - loss: 0.0492 - accuracy: 0.9809 - val_loss: 0.0883 - val_accuracy: 0.9766 - 542ms/epoch - 15ms/step
Validation accuracy for Model of [Learning Rate 0.01 | Num Filters 128 | Dropout Rate 0.3 | FilterSize 7]: 0.9766260385513306
Epoch 1/20
36/36 - 2s - loss: 0.2459 - accuracy: 0.9003 - val_loss: 0.1484 - val_accuracy: 0.9477 - 2s/epoch - 44ms/step
Epoch 2/20
36/36 - 1s - loss: 0.1410 - accuracy: 0.9457 - val_loss: 0.1164 - val_accuracy: 0.9538 - 974ms/epoch - 27ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1193 - accuracy: 0.9523 - val_loss: 0.1351 - val_accuracy: 0.9573 - 1s/epoch - 31ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1135 - accuracy: 0.9542 - val_loss: 0.1141 - val_accuracy: 0.9578 - 977ms/epoch - 27ms/step
Epoch 5/20
36/36 - 1s - loss: 0.1060 - accuracy: 0.9572 - val_loss: 0.0983 - val_accuracy: 0.9588 - 979ms/epoch - 27ms/step
Epoch 6/20
36/36 - 1s - loss: 0.0965 - accuracy: 0.9595 - val_loss: 0.1379 - val_accuracy: 0.9456 - 978ms/epoch - 27ms/step
Epoch 7/20
36/36 - 1s - loss: 0.0952 - accuracy: 0.9620 - val_loss: 0.0917 - val_accuracy: 0.9629 - 661ms/epoch - 18ms/step
Epoch 8/20
36/36 - 1s - loss: 0.0906 - accuracy: 0.9641 - val_loss: 0.0869 - val_accuracy: 0.9685 - 583ms/epoch - 16ms/step
Epoch 9/20
36/36 - 1s - loss: 0.0834 - accuracy: 0.9658 - val_loss: 0.0929 - val_accuracy: 0.9624 - 609ms/epoch - 17ms/step
Epoch 10/20
36/36 - 1s - loss: 0.0844 - accuracy: 0.9647 - val_loss: 0.1262 - val_accuracy: 0.9497 - 616ms/epoch - 17ms/step
Epoch 11/20
36/36 - 1s - loss: 0.0892 - accuracy: 0.9652 - val_loss: 0.1056 - val_accuracy: 0.9563 - 596ms/epoch - 17ms/step
Epoch 12/20
36/36 - 1s - loss: 0.0851 - accuracy: 0.9655 - val_loss: 0.0976 - val_accuracy: 0.9614 - 619ms/epoch - 17ms/step
Epoch 13/20
36/36 - 1s - loss: 0.0820 - accuracy: 0.9675 - val_loss: 0.0967 - val_accuracy: 0.9649 - 590ms/epoch - 16ms/step
Epoch 14/20
36/36 - 1s - loss: 0.0830 - accuracy: 0.9680 - val_loss: 0.0895 - val_accuracy: 0.9665 - 629ms/epoch - 17ms/step
Epoch 15/20
36/36 - 1s - loss: 0.0770 - accuracy: 0.9694 - val_loss: 0.1009 - val_accuracy: 0.9665 - 673ms/epoch - 19ms/step
Epoch 16/20
36/36 - 1s - loss: 0.0767 - accuracy: 0.9692 - val_loss: 0.0828 - val_accuracy: 0.9690 - 713ms/epoch - 20ms/step
Epoch 17/20
36/36 - 1s - loss: 0.0752 - accuracy: 0.9699 - val_loss: 0.0955 - val_accuracy: 0.9644 - 596ms/epoch - 17ms/step
Epoch 18/20
36/36 - 1s - loss: 0.0731 - accuracy: 0.9707 - val_loss: 0.0890 - val_accuracy: 0.9736 - 596ms/epoch - 17ms/step
Epoch 19/20
36/36 - 1s - loss: 0.0772 - accuracy: 0.9690 - val_loss: 0.0904 - val_accuracy: 0.9731 - 592ms/epoch - 16ms/step
Epoch 20/20
36/36 - 1s - loss: 0.0694 - accuracy: 0.9714 - val_loss: 0.0882 - val_accuracy: 0.9700 - 581ms/epoch - 16ms/step
Validation accuracy for Model of [Learning Rate 0.01 | Num Filters 128 | Dropout Rate 0.5 | FilterSize 3]: 0.9700203537940979
Epoch 1/20
36/36 - 2s - loss: 0.2775 - accuracy: 0.8818 - val_loss: 0.1190 - val_accuracy: 0.9563 - 2s/epoch - 45ms/step
Epoch 2/20
36/36 - 1s - loss: 0.1413 - accuracy: 0.9446 - val_loss: 0.1053 - val_accuracy: 0.9604 - 585ms/epoch - 16ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1197 - accuracy: 0.9523 - val_loss: 0.1316 - val_accuracy: 0.9472 - 568ms/epoch - 16ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1084 - accuracy: 0.9550 - val_loss: 0.1168 - val_accuracy: 0.9497 - 603ms/epoch - 17ms/step
Epoch 5/20
36/36 - 1s - loss: 0.0985 - accuracy: 0.9581 - val_loss: 0.1392 - val_accuracy: 0.9466 - 593ms/epoch - 16ms/step
Epoch 6/20
36/36 - 1s - loss: 0.0979 - accuracy: 0.9594 - val_loss: 0.0983 - val_accuracy: 0.9629 - 590ms/epoch - 16ms/step
Epoch 7/20
36/36 - 1s - loss: 0.0899 - accuracy: 0.9635 - val_loss: 0.0882 - val_accuracy: 0.9680 - 561ms/epoch - 16ms/step
Epoch 8/20
36/36 - 1s - loss: 0.0877 - accuracy: 0.9642 - val_loss: 0.1048 - val_accuracy: 0.9609 - 576ms/epoch - 16ms/step
Epoch 9/20
36/36 - 1s - loss: 0.0866 - accuracy: 0.9653 - val_loss: 0.1104 - val_accuracy: 0.9604 - 582ms/epoch - 16ms/step
Epoch 10/20
36/36 - 1s - loss: 0.0839 - accuracy: 0.9671 - val_loss: 0.0990 - val_accuracy: 0.9578 - 595ms/epoch - 17ms/step
Epoch 11/20
36/36 - 1s - loss: 0.0823 - accuracy: 0.9673 - val_loss: 0.0848 - val_accuracy: 0.9721 - 582ms/epoch - 16ms/step
Epoch 12/20
36/36 - 1s - loss: 0.0765 - accuracy: 0.9694 - val_loss: 0.1105 - val_accuracy: 0.9558 - 782ms/epoch - 22ms/step
Epoch 13/20
36/36 - 1s - loss: 0.0788 - accuracy: 0.9674 - val_loss: 0.1025 - val_accuracy: 0.9654 - 1s/epoch - 32ms/step
Epoch 14/20
36/36 - 1s - loss: 0.0819 - accuracy: 0.9673 - val_loss: 0.0912 - val_accuracy: 0.9634 - 993ms/epoch - 28ms/step
Epoch 15/20
36/36 - 1s - loss: 0.0745 - accuracy: 0.9700 - val_loss: 0.1043 - val_accuracy: 0.9660 - 925ms/epoch - 26ms/step
Epoch 16/20
36/36 - 1s - loss: 0.0780 - accuracy: 0.9691 - val_loss: 0.0898 - val_accuracy: 0.9741 - 926ms/epoch - 26ms/step
Epoch 17/20
36/36 - 1s - loss: 0.0751 - accuracy: 0.9660 - val_loss: 0.0935 - val_accuracy: 0.9619 - 756ms/epoch - 21ms/step
Epoch 18/20
36/36 - 1s - loss: 0.0700 - accuracy: 0.9718 - val_loss: 0.1029 - val_accuracy: 0.9644 - 609ms/epoch - 17ms/step
Epoch 19/20
36/36 - 1s - loss: 0.0707 - accuracy: 0.9714 - val_loss: 0.0896 - val_accuracy: 0.9705 - 572ms/epoch - 16ms/step
Epoch 20/20
36/36 - 1s - loss: 0.0692 - accuracy: 0.9730 - val_loss: 0.0920 - val_accuracy: 0.9736 - 574ms/epoch - 16ms/step
Validation accuracy for Model of [Learning Rate 0.01 | Num Filters 128 | Dropout Rate 0.5 | FilterSize 5]: 0.9735772609710693
Epoch 1/20
36/36 - 2s - loss: 0.2655 - accuracy: 0.8814 - val_loss: 0.1476 - val_accuracy: 0.9421 - 2s/epoch - 51ms/step
Epoch 2/20
36/36 - 1s - loss: 0.1335 - accuracy: 0.9473 - val_loss: 0.1265 - val_accuracy: 0.9533 - 872ms/epoch - 24ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1139 - accuracy: 0.9555 - val_loss: 0.1123 - val_accuracy: 0.9558 - 867ms/epoch - 24ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1039 - accuracy: 0.9584 - val_loss: 0.1167 - val_accuracy: 0.9558 - 851ms/epoch - 24ms/step
Epoch 5/20
36/36 - 1s - loss: 0.0939 - accuracy: 0.9622 - val_loss: 0.1265 - val_accuracy: 0.9517 - 832ms/epoch - 23ms/step
Epoch 6/20
36/36 - 1s - loss: 0.0955 - accuracy: 0.9611 - val_loss: 0.1016 - val_accuracy: 0.9604 - 848ms/epoch - 24ms/step
Epoch 7/20
36/36 - 1s - loss: 0.0907 - accuracy: 0.9635 - val_loss: 0.1122 - val_accuracy: 0.9548 - 543ms/epoch - 15ms/step
Epoch 8/20
36/36 - 1s - loss: 0.0903 - accuracy: 0.9644 - val_loss: 0.1113 - val_accuracy: 0.9588 - 513ms/epoch - 14ms/step
Epoch 9/20
36/36 - 1s - loss: 0.0945 - accuracy: 0.9624 - val_loss: 0.1021 - val_accuracy: 0.9619 - 521ms/epoch - 14ms/step
Epoch 10/20
36/36 - 1s - loss: 0.0762 - accuracy: 0.9690 - val_loss: 0.0922 - val_accuracy: 0.9731 - 507ms/epoch - 14ms/step
Epoch 11/20
36/36 - 1s - loss: 0.0788 - accuracy: 0.9698 - val_loss: 0.0894 - val_accuracy: 0.9675 - 552ms/epoch - 15ms/step
Epoch 12/20
36/36 - 1s - loss: 0.0753 - accuracy: 0.9693 - val_loss: 0.1070 - val_accuracy: 0.9629 - 545ms/epoch - 15ms/step
Epoch 13/20
36/36 - 1s - loss: 0.0720 - accuracy: 0.9683 - val_loss: 0.1013 - val_accuracy: 0.9629 - 550ms/epoch - 15ms/step
Epoch 14/20
36/36 - 1s - loss: 0.0735 - accuracy: 0.9696 - val_loss: 0.0869 - val_accuracy: 0.9665 - 508ms/epoch - 14ms/step
Epoch 15/20
36/36 - 1s - loss: 0.0682 - accuracy: 0.9729 - val_loss: 0.1062 - val_accuracy: 0.9609 - 552ms/epoch - 15ms/step
Epoch 16/20
36/36 - 1s - loss: 0.0707 - accuracy: 0.9704 - val_loss: 0.0930 - val_accuracy: 0.9695 - 541ms/epoch - 15ms/step
Epoch 17/20
36/36 - 1s - loss: 0.0691 - accuracy: 0.9734 - val_loss: 0.1152 - val_accuracy: 0.9614 - 559ms/epoch - 16ms/step
Epoch 18/20
36/36 - 1s - loss: 0.0670 - accuracy: 0.9727 - val_loss: 0.1041 - val_accuracy: 0.9634 - 511ms/epoch - 14ms/step
Epoch 19/20
36/36 - 1s - loss: 0.0649 - accuracy: 0.9734 - val_loss: 0.1113 - val_accuracy: 0.9543 - 600ms/epoch - 17ms/step
Epoch 20/20
36/36 - 1s - loss: 0.0611 - accuracy: 0.9724 - val_loss: 0.0896 - val_accuracy: 0.9726 - 633ms/epoch - 18ms/step
Validation accuracy for Model of [Learning Rate 0.01 | Num Filters 128 | Dropout Rate 0.5 | FilterSize 7]: 0.9725610017776489
Epoch 1/20
36/36 - 2s - loss: 0.2885 - accuracy: 0.8823 - val_loss: 0.1682 - val_accuracy: 0.9411 - 2s/epoch - 45ms/step
Epoch 2/20
36/36 - 1s - loss: 0.1706 - accuracy: 0.9342 - val_loss: 0.1399 - val_accuracy: 0.9461 - 759ms/epoch - 21ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1459 - accuracy: 0.9422 - val_loss: 0.1242 - val_accuracy: 0.9502 - 985ms/epoch - 27ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1323 - accuracy: 0.9478 - val_loss: 0.1089 - val_accuracy: 0.9522 - 962ms/epoch - 27ms/step
Epoch 5/20
36/36 - 1s - loss: 0.1261 - accuracy: 0.9488 - val_loss: 0.1404 - val_accuracy: 0.9380 - 981ms/epoch - 27ms/step
Epoch 6/20
36/36 - 1s - loss: 0.1216 - accuracy: 0.9496 - val_loss: 0.1193 - val_accuracy: 0.9533 - 969ms/epoch - 27ms/step
Epoch 7/20
36/36 - 1s - loss: 0.1183 - accuracy: 0.9506 - val_loss: 0.1367 - val_accuracy: 0.9446 - 812ms/epoch - 23ms/step
Epoch 8/20
36/36 - 1s - loss: 0.1189 - accuracy: 0.9524 - val_loss: 0.1063 - val_accuracy: 0.9573 - 590ms/epoch - 16ms/step
Epoch 9/20
36/36 - 1s - loss: 0.1186 - accuracy: 0.9539 - val_loss: 0.1017 - val_accuracy: 0.9639 - 600ms/epoch - 17ms/step
Epoch 10/20
36/36 - 1s - loss: 0.1139 - accuracy: 0.9546 - val_loss: 0.1050 - val_accuracy: 0.9614 - 585ms/epoch - 16ms/step
Epoch 11/20
36/36 - 1s - loss: 0.1079 - accuracy: 0.9532 - val_loss: 0.0915 - val_accuracy: 0.9715 - 631ms/epoch - 18ms/step
Epoch 12/20
36/36 - 1s - loss: 0.1032 - accuracy: 0.9546 - val_loss: 0.1104 - val_accuracy: 0.9573 - 617ms/epoch - 17ms/step
Epoch 13/20
36/36 - 1s - loss: 0.1033 - accuracy: 0.9566 - val_loss: 0.0928 - val_accuracy: 0.9670 - 726ms/epoch - 20ms/step
Epoch 14/20
36/36 - 1s - loss: 0.1055 - accuracy: 0.9570 - val_loss: 0.1022 - val_accuracy: 0.9609 - 631ms/epoch - 18ms/step
Epoch 15/20
36/36 - 1s - loss: 0.1005 - accuracy: 0.9574 - val_loss: 0.1105 - val_accuracy: 0.9568 - 610ms/epoch - 17ms/step
Epoch 16/20
36/36 - 1s - loss: 0.0986 - accuracy: 0.9595 - val_loss: 0.1074 - val_accuracy: 0.9588 - 713ms/epoch - 20ms/step
Epoch 17/20
36/36 - 1s - loss: 0.0976 - accuracy: 0.9600 - val_loss: 0.1069 - val_accuracy: 0.9578 - 650ms/epoch - 18ms/step
Epoch 18/20
36/36 - 1s - loss: 0.1019 - accuracy: 0.9583 - val_loss: 0.1115 - val_accuracy: 0.9492 - 582ms/epoch - 16ms/step
Epoch 19/20
36/36 - 1s - loss: 0.0987 - accuracy: 0.9595 - val_loss: 0.0995 - val_accuracy: 0.9660 - 572ms/epoch - 16ms/step
Epoch 20/20
36/36 - 1s - loss: 0.0920 - accuracy: 0.9627 - val_loss: 0.1004 - val_accuracy: 0.9649 - 618ms/epoch - 17ms/step
Validation accuracy for Model of [Learning Rate 0.01 | Num Filters 128 | Dropout Rate 0.7 | FilterSize 3]: 0.9649389982223511
Epoch 1/20
36/36 - 2s - loss: 0.2836 - accuracy: 0.8759 - val_loss: 0.1337 - val_accuracy: 0.9507 - 2s/epoch - 44ms/step
Epoch 2/20
36/36 - 1s - loss: 0.1589 - accuracy: 0.9408 - val_loss: 0.1227 - val_accuracy: 0.9527 - 906ms/epoch - 25ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1437 - accuracy: 0.9453 - val_loss: 0.1623 - val_accuracy: 0.9375 - 927ms/epoch - 26ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1297 - accuracy: 0.9493 - val_loss: 0.1248 - val_accuracy: 0.9507 - 918ms/epoch - 26ms/step
Epoch 5/20
36/36 - 1s - loss: 0.1203 - accuracy: 0.9522 - val_loss: 0.0976 - val_accuracy: 0.9604 - 922ms/epoch - 26ms/step
Epoch 6/20
36/36 - 1s - loss: 0.1212 - accuracy: 0.9531 - val_loss: 0.1026 - val_accuracy: 0.9629 - 937ms/epoch - 26ms/step
Epoch 7/20
36/36 - 1s - loss: 0.1106 - accuracy: 0.9535 - val_loss: 0.1003 - val_accuracy: 0.9614 - 603ms/epoch - 17ms/step
Epoch 8/20
36/36 - 1s - loss: 0.1053 - accuracy: 0.9548 - val_loss: 0.1053 - val_accuracy: 0.9599 - 588ms/epoch - 16ms/step
Epoch 9/20
36/36 - 1s - loss: 0.1030 - accuracy: 0.9568 - val_loss: 0.1057 - val_accuracy: 0.9573 - 597ms/epoch - 17ms/step
Epoch 10/20
36/36 - 1s - loss: 0.1006 - accuracy: 0.9593 - val_loss: 0.1014 - val_accuracy: 0.9593 - 613ms/epoch - 17ms/step
Epoch 11/20
36/36 - 1s - loss: 0.1047 - accuracy: 0.9564 - val_loss: 0.0976 - val_accuracy: 0.9614 - 634ms/epoch - 18ms/step
Epoch 12/20
36/36 - 1s - loss: 0.0970 - accuracy: 0.9608 - val_loss: 0.1096 - val_accuracy: 0.9573 - 765ms/epoch - 21ms/step
Epoch 13/20
36/36 - 1s - loss: 0.0951 - accuracy: 0.9619 - val_loss: 0.0976 - val_accuracy: 0.9660 - 608ms/epoch - 17ms/step
Epoch 14/20
36/36 - 1s - loss: 0.0943 - accuracy: 0.9598 - val_loss: 0.1077 - val_accuracy: 0.9624 - 573ms/epoch - 16ms/step
Epoch 15/20
36/36 - 1s - loss: 0.0972 - accuracy: 0.9599 - val_loss: 0.1040 - val_accuracy: 0.9660 - 617ms/epoch - 17ms/step
Epoch 16/20
36/36 - 1s - loss: 0.1003 - accuracy: 0.9592 - val_loss: 0.1242 - val_accuracy: 0.9487 - 572ms/epoch - 16ms/step
Epoch 17/20
36/36 - 1s - loss: 0.0939 - accuracy: 0.9603 - val_loss: 0.1114 - val_accuracy: 0.9599 - 567ms/epoch - 16ms/step
Epoch 18/20
36/36 - 1s - loss: 0.0871 - accuracy: 0.9625 - val_loss: 0.0978 - val_accuracy: 0.9721 - 573ms/epoch - 16ms/step
Epoch 19/20
36/36 - 1s - loss: 0.0886 - accuracy: 0.9604 - val_loss: 0.0987 - val_accuracy: 0.9654 - 578ms/epoch - 16ms/step
Epoch 20/20
36/36 - 1s - loss: 0.0898 - accuracy: 0.9615 - val_loss: 0.1135 - val_accuracy: 0.9527 - 576ms/epoch - 16ms/step
Validation accuracy for Model of [Learning Rate 0.01 | Num Filters 128 | Dropout Rate 0.7 | FilterSize 5]: 0.9527438879013062
Epoch 1/20
36/36 - 2s - loss: 0.2697 - accuracy: 0.8873 - val_loss: 0.1344 - val_accuracy: 0.9502 - 2s/epoch - 42ms/step
Epoch 2/20
36/36 - 1s - loss: 0.1599 - accuracy: 0.9393 - val_loss: 0.1114 - val_accuracy: 0.9563 - 861ms/epoch - 24ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1291 - accuracy: 0.9493 - val_loss: 0.1129 - val_accuracy: 0.9512 - 849ms/epoch - 24ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1203 - accuracy: 0.9534 - val_loss: 0.0978 - val_accuracy: 0.9624 - 854ms/epoch - 24ms/step
Epoch 5/20
36/36 - 1s - loss: 0.1181 - accuracy: 0.9540 - val_loss: 0.1098 - val_accuracy: 0.9588 - 855ms/epoch - 24ms/step
Epoch 6/20
36/36 - 1s - loss: 0.1142 - accuracy: 0.9542 - val_loss: 0.1061 - val_accuracy: 0.9568 - 836ms/epoch - 23ms/step
Epoch 7/20
36/36 - 1s - loss: 0.1052 - accuracy: 0.9562 - val_loss: 0.1176 - val_accuracy: 0.9522 - 704ms/epoch - 20ms/step
Epoch 8/20
36/36 - 1s - loss: 0.0964 - accuracy: 0.9606 - val_loss: 0.1342 - val_accuracy: 0.9421 - 551ms/epoch - 15ms/step
Epoch 9/20
36/36 - 1s - loss: 0.0944 - accuracy: 0.9606 - val_loss: 0.1118 - val_accuracy: 0.9487 - 546ms/epoch - 15ms/step
Epoch 10/20
36/36 - 1s - loss: 0.0969 - accuracy: 0.9594 - val_loss: 0.1014 - val_accuracy: 0.9604 - 550ms/epoch - 15ms/step
Epoch 11/20
36/36 - 1s - loss: 0.0951 - accuracy: 0.9613 - val_loss: 0.1062 - val_accuracy: 0.9548 - 500ms/epoch - 14ms/step
Epoch 12/20
36/36 - 1s - loss: 0.0914 - accuracy: 0.9643 - val_loss: 0.1274 - val_accuracy: 0.9568 - 540ms/epoch - 15ms/step
Epoch 13/20
36/36 - 1s - loss: 0.0939 - accuracy: 0.9580 - val_loss: 0.1040 - val_accuracy: 0.9609 - 514ms/epoch - 14ms/step
Epoch 14/20
36/36 - 1s - loss: 0.0909 - accuracy: 0.9618 - val_loss: 0.1230 - val_accuracy: 0.9533 - 519ms/epoch - 14ms/step
Epoch 15/20
36/36 - 1s - loss: 0.0977 - accuracy: 0.9602 - val_loss: 0.0997 - val_accuracy: 0.9568 - 531ms/epoch - 15ms/step
Epoch 16/20
36/36 - 1s - loss: 0.0846 - accuracy: 0.9646 - val_loss: 0.0990 - val_accuracy: 0.9654 - 514ms/epoch - 14ms/step
Epoch 17/20
36/36 - 1s - loss: 0.0823 - accuracy: 0.9645 - val_loss: 0.1208 - val_accuracy: 0.9512 - 535ms/epoch - 15ms/step
Epoch 18/20
36/36 - 1s - loss: 0.0881 - accuracy: 0.9623 - val_loss: 0.1077 - val_accuracy: 0.9609 - 549ms/epoch - 15ms/step
Epoch 19/20
36/36 - 1s - loss: 0.0821 - accuracy: 0.9638 - val_loss: 0.0993 - val_accuracy: 0.9614 - 500ms/epoch - 14ms/step
Epoch 20/20
36/36 - 1s - loss: 0.0821 - accuracy: 0.9635 - val_loss: 0.1030 - val_accuracy: 0.9634 - 537ms/epoch - 15ms/step
Validation accuracy for Model of [Learning Rate 0.01 | Num Filters 128 | Dropout Rate 0.7 | FilterSize 7]: 0.9634146094322205
Epoch 1/20
36/36 - 1s - loss: 0.8639 - accuracy: 0.7784 - val_loss: 0.2070 - val_accuracy: 0.9228 - 1s/epoch - 34ms/step
Epoch 2/20
36/36 - 0s - loss: 0.1554 - accuracy: 0.9335 - val_loss: 0.1580 - val_accuracy: 0.9421 - 343ms/epoch - 10ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1457 - accuracy: 0.9373 - val_loss: 0.1369 - val_accuracy: 0.9395 - 322ms/epoch - 9ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1409 - accuracy: 0.9405 - val_loss: 0.1306 - val_accuracy: 0.9512 - 280ms/epoch - 8ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1304 - accuracy: 0.9423 - val_loss: 0.1192 - val_accuracy: 0.9593 - 329ms/epoch - 9ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1342 - accuracy: 0.9375 - val_loss: 0.1510 - val_accuracy: 0.9538 - 361ms/epoch - 10ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1250 - accuracy: 0.9407 - val_loss: 0.1167 - val_accuracy: 0.9375 - 429ms/epoch - 12ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1194 - accuracy: 0.9440 - val_loss: 0.1205 - val_accuracy: 0.9446 - 447ms/epoch - 12ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1194 - accuracy: 0.9447 - val_loss: 0.1068 - val_accuracy: 0.9466 - 446ms/epoch - 12ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1175 - accuracy: 0.9443 - val_loss: 0.1245 - val_accuracy: 0.9385 - 483ms/epoch - 13ms/step
Epoch 11/20
36/36 - 0s - loss: 0.1218 - accuracy: 0.9407 - val_loss: 0.1244 - val_accuracy: 0.9405 - 452ms/epoch - 13ms/step
Epoch 12/20
36/36 - 0s - loss: 0.1243 - accuracy: 0.9426 - val_loss: 0.1157 - val_accuracy: 0.9472 - 467ms/epoch - 13ms/step
Epoch 13/20
36/36 - 0s - loss: 0.1102 - accuracy: 0.9520 - val_loss: 0.1088 - val_accuracy: 0.9456 - 459ms/epoch - 13ms/step
Epoch 14/20
36/36 - 1s - loss: 0.1084 - accuracy: 0.9502 - val_loss: 0.1179 - val_accuracy: 0.9416 - 538ms/epoch - 15ms/step
Epoch 15/20
36/36 - 1s - loss: 0.1174 - accuracy: 0.9480 - val_loss: 0.1494 - val_accuracy: 0.9345 - 577ms/epoch - 16ms/step
Epoch 16/20
36/36 - 0s - loss: 0.1226 - accuracy: 0.9434 - val_loss: 0.1197 - val_accuracy: 0.9451 - 406ms/epoch - 11ms/step
Epoch 17/20
36/36 - 0s - loss: 0.1368 - accuracy: 0.9327 - val_loss: 0.1455 - val_accuracy: 0.9619 - 292ms/epoch - 8ms/step
Epoch 18/20
36/36 - 0s - loss: 0.1239 - accuracy: 0.9390 - val_loss: 0.1240 - val_accuracy: 0.9624 - 281ms/epoch - 8ms/step
Epoch 19/20
36/36 - 0s - loss: 0.1125 - accuracy: 0.9415 - val_loss: 0.1242 - val_accuracy: 0.9588 - 298ms/epoch - 8ms/step
Epoch 20/20
36/36 - 0s - loss: 0.1221 - accuracy: 0.9430 - val_loss: 0.1376 - val_accuracy: 0.9609 - 294ms/epoch - 8ms/step
Validation accuracy for Model of [Learning Rate 0.1 | Num Filters 32 | Dropout Rate 0.3 | FilterSize 3]: 0.9608739614486694
Epoch 1/20
36/36 - 1s - loss: 0.7538 - accuracy: 0.7979 - val_loss: 0.2086 - val_accuracy: 0.9207 - 1s/epoch - 34ms/step
Epoch 2/20
36/36 - 0s - loss: 0.1819 - accuracy: 0.9264 - val_loss: 0.1304 - val_accuracy: 0.9522 - 378ms/epoch - 10ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1369 - accuracy: 0.9466 - val_loss: 0.1381 - val_accuracy: 0.9512 - 396ms/epoch - 11ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1324 - accuracy: 0.9480 - val_loss: 0.1111 - val_accuracy: 0.9578 - 322ms/epoch - 9ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1275 - accuracy: 0.9483 - val_loss: 0.1297 - val_accuracy: 0.9578 - 277ms/epoch - 8ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1315 - accuracy: 0.9459 - val_loss: 0.1320 - val_accuracy: 0.9487 - 291ms/epoch - 8ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1159 - accuracy: 0.9533 - val_loss: 0.1404 - val_accuracy: 0.9583 - 275ms/epoch - 8ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1143 - accuracy: 0.9539 - val_loss: 0.1097 - val_accuracy: 0.9604 - 288ms/epoch - 8ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1029 - accuracy: 0.9590 - val_loss: 0.1190 - val_accuracy: 0.9573 - 336ms/epoch - 9ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1028 - accuracy: 0.9581 - val_loss: 0.1290 - val_accuracy: 0.9538 - 288ms/epoch - 8ms/step
Epoch 11/20
36/36 - 0s - loss: 0.1124 - accuracy: 0.9551 - val_loss: 0.1180 - val_accuracy: 0.9568 - 290ms/epoch - 8ms/step
Epoch 12/20
36/36 - 0s - loss: 0.1017 - accuracy: 0.9589 - val_loss: 0.1239 - val_accuracy: 0.9599 - 277ms/epoch - 8ms/step
Epoch 13/20
36/36 - 0s - loss: 0.1124 - accuracy: 0.9538 - val_loss: 0.1022 - val_accuracy: 0.9710 - 338ms/epoch - 9ms/step
Epoch 14/20
36/36 - 0s - loss: 0.1118 - accuracy: 0.9555 - val_loss: 0.0941 - val_accuracy: 0.9705 - 334ms/epoch - 9ms/step
Epoch 15/20
36/36 - 0s - loss: 0.1064 - accuracy: 0.9567 - val_loss: 0.1364 - val_accuracy: 0.9573 - 281ms/epoch - 8ms/step
Epoch 16/20
36/36 - 0s - loss: 0.1018 - accuracy: 0.9580 - val_loss: 0.1020 - val_accuracy: 0.9624 - 460ms/epoch - 13ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0946 - accuracy: 0.9602 - val_loss: 0.1604 - val_accuracy: 0.9614 - 460ms/epoch - 13ms/step
Epoch 18/20
36/36 - 1s - loss: 0.1055 - accuracy: 0.9572 - val_loss: 0.1082 - val_accuracy: 0.9639 - 539ms/epoch - 15ms/step
Epoch 19/20
36/36 - 1s - loss: 0.0973 - accuracy: 0.9634 - val_loss: 0.1128 - val_accuracy: 0.9639 - 559ms/epoch - 16ms/step
Epoch 20/20
36/36 - 1s - loss: 0.1026 - accuracy: 0.9564 - val_loss: 0.1391 - val_accuracy: 0.9604 - 565ms/epoch - 16ms/step
Validation accuracy for Model of [Learning Rate 0.1 | Num Filters 32 | Dropout Rate 0.3 | FilterSize 5]: 0.9603658318519592
Epoch 1/20
36/36 - 2s - loss: 1.5714 - accuracy: 0.7252 - val_loss: 0.2476 - val_accuracy: 0.9157 - 2s/epoch - 44ms/step
Epoch 2/20
36/36 - 0s - loss: 0.1875 - accuracy: 0.9228 - val_loss: 0.1403 - val_accuracy: 0.9456 - 313ms/epoch - 9ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1493 - accuracy: 0.9348 - val_loss: 0.1575 - val_accuracy: 0.9522 - 290ms/epoch - 8ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1382 - accuracy: 0.9357 - val_loss: 0.1242 - val_accuracy: 0.9533 - 279ms/epoch - 8ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1246 - accuracy: 0.9383 - val_loss: 0.1277 - val_accuracy: 0.9553 - 279ms/epoch - 8ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1252 - accuracy: 0.9387 - val_loss: 0.1344 - val_accuracy: 0.9355 - 276ms/epoch - 8ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1251 - accuracy: 0.9373 - val_loss: 0.1572 - val_accuracy: 0.9456 - 280ms/epoch - 8ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1198 - accuracy: 0.9442 - val_loss: 0.1377 - val_accuracy: 0.9380 - 319ms/epoch - 9ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1205 - accuracy: 0.9406 - val_loss: 0.1389 - val_accuracy: 0.9278 - 319ms/epoch - 9ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1174 - accuracy: 0.9449 - val_loss: 0.1463 - val_accuracy: 0.9141 - 293ms/epoch - 8ms/step
Epoch 11/20
36/36 - 0s - loss: 0.1178 - accuracy: 0.9420 - val_loss: 0.1286 - val_accuracy: 0.9426 - 267ms/epoch - 7ms/step
Epoch 12/20
36/36 - 0s - loss: 0.1119 - accuracy: 0.9453 - val_loss: 0.1422 - val_accuracy: 0.9223 - 279ms/epoch - 8ms/step
Epoch 13/20
36/36 - 0s - loss: 0.1113 - accuracy: 0.9469 - val_loss: 0.1465 - val_accuracy: 0.9299 - 326ms/epoch - 9ms/step
Epoch 14/20
36/36 - 0s - loss: 0.1118 - accuracy: 0.9440 - val_loss: 0.1382 - val_accuracy: 0.9294 - 310ms/epoch - 9ms/step
Epoch 15/20
36/36 - 0s - loss: 0.1106 - accuracy: 0.9441 - val_loss: 0.1470 - val_accuracy: 0.9339 - 365ms/epoch - 10ms/step
Epoch 16/20
36/36 - 0s - loss: 0.1082 - accuracy: 0.9471 - val_loss: 0.1220 - val_accuracy: 0.9329 - 346ms/epoch - 10ms/step
Epoch 17/20
36/36 - 0s - loss: 0.1115 - accuracy: 0.9438 - val_loss: 0.1348 - val_accuracy: 0.9314 - 284ms/epoch - 8ms/step
Epoch 18/20
36/36 - 0s - loss: 0.1130 - accuracy: 0.9471 - val_loss: 0.1753 - val_accuracy: 0.9141 - 271ms/epoch - 8ms/step
Epoch 19/20
36/36 - 0s - loss: 0.1293 - accuracy: 0.9396 - val_loss: 0.1496 - val_accuracy: 0.9390 - 274ms/epoch - 8ms/step
Epoch 20/20
36/36 - 0s - loss: 0.1208 - accuracy: 0.9462 - val_loss: 0.1313 - val_accuracy: 0.9253 - 283ms/epoch - 8ms/step
Validation accuracy for Model of [Learning Rate 0.1 | Num Filters 32 | Dropout Rate 0.3 | FilterSize 7]: 0.9253048896789551
Epoch 1/20
36/36 - 2s - loss: 0.6708 - accuracy: 0.7854 - val_loss: 0.2244 - val_accuracy: 0.9106 - 2s/epoch - 52ms/step
Epoch 2/20
36/36 - 0s - loss: 0.2067 - accuracy: 0.9153 - val_loss: 0.1683 - val_accuracy: 0.9370 - 482ms/epoch - 13ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1953 - accuracy: 0.9226 - val_loss: 0.1886 - val_accuracy: 0.9060 - 440ms/epoch - 12ms/step
Epoch 4/20
36/36 - 0s - loss: 0.2123 - accuracy: 0.9150 - val_loss: 0.1951 - val_accuracy: 0.9009 - 498ms/epoch - 14ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1999 - accuracy: 0.9135 - val_loss: 0.1488 - val_accuracy: 0.9309 - 481ms/epoch - 13ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1859 - accuracy: 0.9176 - val_loss: 0.1413 - val_accuracy: 0.9309 - 328ms/epoch - 9ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1711 - accuracy: 0.9264 - val_loss: 0.1487 - val_accuracy: 0.9248 - 346ms/epoch - 10ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1714 - accuracy: 0.9211 - val_loss: 0.1512 - val_accuracy: 0.9319 - 302ms/epoch - 8ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1716 - accuracy: 0.9236 - val_loss: 0.1319 - val_accuracy: 0.9324 - 326ms/epoch - 9ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1576 - accuracy: 0.9303 - val_loss: 0.1426 - val_accuracy: 0.9324 - 300ms/epoch - 8ms/step
Epoch 11/20
36/36 - 0s - loss: 0.1579 - accuracy: 0.9317 - val_loss: 0.1494 - val_accuracy: 0.9157 - 299ms/epoch - 8ms/step
Epoch 12/20
36/36 - 0s - loss: 0.1517 - accuracy: 0.9350 - val_loss: 0.1568 - val_accuracy: 0.9273 - 306ms/epoch - 8ms/step
Epoch 13/20
36/36 - 0s - loss: 0.1595 - accuracy: 0.9331 - val_loss: 0.1412 - val_accuracy: 0.9319 - 289ms/epoch - 8ms/step
Epoch 14/20
36/36 - 0s - loss: 0.1549 - accuracy: 0.9338 - val_loss: 0.1307 - val_accuracy: 0.9212 - 342ms/epoch - 9ms/step
Epoch 15/20
36/36 - 0s - loss: 0.1601 - accuracy: 0.9318 - val_loss: 0.1666 - val_accuracy: 0.9157 - 341ms/epoch - 9ms/step
Epoch 16/20
36/36 - 0s - loss: 0.1546 - accuracy: 0.9333 - val_loss: 0.1557 - val_accuracy: 0.9319 - 294ms/epoch - 8ms/step
Epoch 17/20
36/36 - 0s - loss: 0.1654 - accuracy: 0.9300 - val_loss: 0.1318 - val_accuracy: 0.9380 - 289ms/epoch - 8ms/step
Epoch 18/20
36/36 - 0s - loss: 0.1513 - accuracy: 0.9378 - val_loss: 0.1642 - val_accuracy: 0.9101 - 349ms/epoch - 10ms/step
Epoch 19/20
36/36 - 0s - loss: 0.1561 - accuracy: 0.9336 - val_loss: 0.1590 - val_accuracy: 0.9268 - 290ms/epoch - 8ms/step
Epoch 20/20
36/36 - 0s - loss: 0.1632 - accuracy: 0.9272 - val_loss: 0.1531 - val_accuracy: 0.9202 - 339ms/epoch - 9ms/step
Validation accuracy for Model of [Learning Rate 0.1 | Num Filters 32 | Dropout Rate 0.5 | FilterSize 3]: 0.920223593711853
Epoch 1/20
36/36 - 1s - loss: 0.8832 - accuracy: 0.7802 - val_loss: 0.2590 - val_accuracy: 0.8211 - 1s/epoch - 36ms/step
Epoch 2/20
36/36 - 0s - loss: 0.2042 - accuracy: 0.9099 - val_loss: 0.1808 - val_accuracy: 0.9416 - 378ms/epoch - 11ms/step
Epoch 3/20
36/36 - 0s - loss: 0.2061 - accuracy: 0.9199 - val_loss: 0.1977 - val_accuracy: 0.9116 - 357ms/epoch - 10ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1747 - accuracy: 0.9253 - val_loss: 0.1724 - val_accuracy: 0.9380 - 401ms/epoch - 11ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1643 - accuracy: 0.9295 - val_loss: 0.1148 - val_accuracy: 0.9558 - 333ms/epoch - 9ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1685 - accuracy: 0.9301 - val_loss: 0.1779 - val_accuracy: 0.9426 - 295ms/epoch - 8ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1707 - accuracy: 0.9299 - val_loss: 0.1628 - val_accuracy: 0.9527 - 307ms/epoch - 9ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1649 - accuracy: 0.9318 - val_loss: 0.1434 - val_accuracy: 0.9461 - 286ms/epoch - 8ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1691 - accuracy: 0.9259 - val_loss: 0.1447 - val_accuracy: 0.9426 - 328ms/epoch - 9ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1521 - accuracy: 0.9315 - val_loss: 0.1574 - val_accuracy: 0.9421 - 284ms/epoch - 8ms/step
Epoch 11/20
36/36 - 0s - loss: 0.1436 - accuracy: 0.9342 - val_loss: 0.1429 - val_accuracy: 0.9548 - 289ms/epoch - 8ms/step
Epoch 12/20
36/36 - 0s - loss: 0.1510 - accuracy: 0.9297 - val_loss: 0.1615 - val_accuracy: 0.9599 - 301ms/epoch - 8ms/step
Epoch 13/20
36/36 - 0s - loss: 0.1615 - accuracy: 0.9283 - val_loss: 0.1741 - val_accuracy: 0.9446 - 394ms/epoch - 11ms/step
Epoch 14/20
36/36 - 0s - loss: 0.1493 - accuracy: 0.9334 - val_loss: 0.1331 - val_accuracy: 0.9533 - 479ms/epoch - 13ms/step
Epoch 15/20
36/36 - 0s - loss: 0.1528 - accuracy: 0.9332 - val_loss: 0.1658 - val_accuracy: 0.9502 - 447ms/epoch - 12ms/step
Epoch 16/20
36/36 - 1s - loss: 0.1528 - accuracy: 0.9346 - val_loss: 0.1697 - val_accuracy: 0.9497 - 521ms/epoch - 14ms/step
Epoch 17/20
36/36 - 1s - loss: 0.1558 - accuracy: 0.9324 - val_loss: 0.2129 - val_accuracy: 0.9538 - 540ms/epoch - 15ms/step
Epoch 18/20
36/36 - 1s - loss: 0.1733 - accuracy: 0.9288 - val_loss: 0.2084 - val_accuracy: 0.9421 - 521ms/epoch - 14ms/step
Epoch 19/20
36/36 - 0s - loss: 0.1550 - accuracy: 0.9334 - val_loss: 0.1440 - val_accuracy: 0.9583 - 447ms/epoch - 12ms/step
Epoch 20/20
36/36 - 0s - loss: 0.1497 - accuracy: 0.9325 - val_loss: 0.1501 - val_accuracy: 0.9522 - 486ms/epoch - 13ms/step
Validation accuracy for Model of [Learning Rate 0.1 | Num Filters 32 | Dropout Rate 0.5 | FilterSize 5]: 0.952235758304596
Epoch 1/20
36/36 - 1s - loss: 1.0778 - accuracy: 0.7635 - val_loss: 0.2120 - val_accuracy: 0.9304 - 1s/epoch - 41ms/step
Epoch 2/20
36/36 - 0s - loss: 0.2071 - accuracy: 0.9092 - val_loss: 0.1484 - val_accuracy: 0.9533 - 365ms/epoch - 10ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1707 - accuracy: 0.9224 - val_loss: 0.1395 - val_accuracy: 0.9548 - 374ms/epoch - 10ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1703 - accuracy: 0.9194 - val_loss: 0.1598 - val_accuracy: 0.9538 - 347ms/epoch - 10ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1710 - accuracy: 0.9226 - val_loss: 0.1645 - val_accuracy: 0.9512 - 312ms/epoch - 9ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1605 - accuracy: 0.9278 - val_loss: 0.1293 - val_accuracy: 0.9578 - 272ms/epoch - 8ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1512 - accuracy: 0.9267 - val_loss: 0.1279 - val_accuracy: 0.9548 - 286ms/epoch - 8ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1400 - accuracy: 0.9318 - val_loss: 0.1352 - val_accuracy: 0.9563 - 319ms/epoch - 9ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1437 - accuracy: 0.9319 - val_loss: 0.1160 - val_accuracy: 0.9558 - 281ms/epoch - 8ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1342 - accuracy: 0.9336 - val_loss: 0.1273 - val_accuracy: 0.9654 - 279ms/epoch - 8ms/step
Epoch 11/20
36/36 - 0s - loss: 0.1369 - accuracy: 0.9315 - val_loss: 0.1194 - val_accuracy: 0.9593 - 263ms/epoch - 7ms/step
Epoch 12/20
36/36 - 0s - loss: 0.1457 - accuracy: 0.9286 - val_loss: 0.1056 - val_accuracy: 0.9629 - 326ms/epoch - 9ms/step
Epoch 13/20
36/36 - 0s - loss: 0.1362 - accuracy: 0.9328 - val_loss: 0.1510 - val_accuracy: 0.9284 - 275ms/epoch - 8ms/step
Epoch 14/20
36/36 - 0s - loss: 0.1284 - accuracy: 0.9360 - val_loss: 0.1484 - val_accuracy: 0.9258 - 278ms/epoch - 8ms/step
Epoch 15/20
36/36 - 0s - loss: 0.1325 - accuracy: 0.9307 - val_loss: 0.1213 - val_accuracy: 0.9629 - 273ms/epoch - 8ms/step
Epoch 16/20
36/36 - 0s - loss: 0.1332 - accuracy: 0.9318 - val_loss: 0.1927 - val_accuracy: 0.8882 - 266ms/epoch - 7ms/step
Epoch 17/20
36/36 - 0s - loss: 0.1369 - accuracy: 0.9317 - val_loss: 0.1397 - val_accuracy: 0.9329 - 273ms/epoch - 8ms/step
Epoch 18/20
36/36 - 0s - loss: 0.1466 - accuracy: 0.9305 - val_loss: 0.1554 - val_accuracy: 0.9157 - 267ms/epoch - 7ms/step
Epoch 19/20
36/36 - 0s - loss: 0.1644 - accuracy: 0.9076 - val_loss: 0.1543 - val_accuracy: 0.9583 - 284ms/epoch - 8ms/step
Epoch 20/20
36/36 - 0s - loss: 0.1570 - accuracy: 0.9191 - val_loss: 0.1358 - val_accuracy: 0.9253 - 270ms/epoch - 7ms/step
Validation accuracy for Model of [Learning Rate 0.1 | Num Filters 32 | Dropout Rate 0.5 | FilterSize 7]: 0.9253048896789551
Epoch 1/20
36/36 - 1s - loss: 0.5231 - accuracy: 0.8340 - val_loss: 0.1594 - val_accuracy: 0.9400 - 1s/epoch - 35ms/step
Epoch 2/20
36/36 - 0s - loss: 0.1996 - accuracy: 0.9249 - val_loss: 0.1559 - val_accuracy: 0.9477 - 293ms/epoch - 8ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1958 - accuracy: 0.9255 - val_loss: 0.1410 - val_accuracy: 0.9487 - 288ms/epoch - 8ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1971 - accuracy: 0.9265 - val_loss: 0.1730 - val_accuracy: 0.9350 - 292ms/epoch - 8ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1838 - accuracy: 0.9251 - val_loss: 0.1387 - val_accuracy: 0.9482 - 290ms/epoch - 8ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1837 - accuracy: 0.9276 - val_loss: 0.1562 - val_accuracy: 0.9477 - 301ms/epoch - 8ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1795 - accuracy: 0.9294 - val_loss: 0.1535 - val_accuracy: 0.9446 - 296ms/epoch - 8ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1882 - accuracy: 0.9236 - val_loss: 0.1573 - val_accuracy: 0.9502 - 289ms/epoch - 8ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1924 - accuracy: 0.9234 - val_loss: 0.1722 - val_accuracy: 0.9466 - 408ms/epoch - 11ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1943 - accuracy: 0.9218 - val_loss: 0.1728 - val_accuracy: 0.9421 - 457ms/epoch - 13ms/step
Epoch 11/20
36/36 - 0s - loss: 0.1895 - accuracy: 0.9210 - val_loss: 0.1387 - val_accuracy: 0.9578 - 463ms/epoch - 13ms/step
Epoch 12/20
36/36 - 0s - loss: 0.1854 - accuracy: 0.9242 - val_loss: 0.1345 - val_accuracy: 0.9522 - 464ms/epoch - 13ms/step
Epoch 13/20
36/36 - 0s - loss: 0.1898 - accuracy: 0.9234 - val_loss: 0.1563 - val_accuracy: 0.9527 - 469ms/epoch - 13ms/step
Epoch 14/20
36/36 - 0s - loss: 0.1985 - accuracy: 0.9257 - val_loss: 0.1781 - val_accuracy: 0.9517 - 470ms/epoch - 13ms/step
Epoch 15/20
36/36 - 0s - loss: 0.1963 - accuracy: 0.9203 - val_loss: 0.1935 - val_accuracy: 0.9426 - 463ms/epoch - 13ms/step
Epoch 16/20
36/36 - 0s - loss: 0.1811 - accuracy: 0.9271 - val_loss: 0.1471 - val_accuracy: 0.9553 - 453ms/epoch - 13ms/step
Epoch 17/20
36/36 - 0s - loss: 0.1925 - accuracy: 0.9232 - val_loss: 0.1527 - val_accuracy: 0.9553 - 424ms/epoch - 12ms/step
Epoch 18/20
36/36 - 0s - loss: 0.2007 - accuracy: 0.9151 - val_loss: 0.1784 - val_accuracy: 0.9502 - 469ms/epoch - 13ms/step
Epoch 19/20
36/36 - 0s - loss: 0.1977 - accuracy: 0.9134 - val_loss: 0.1519 - val_accuracy: 0.9558 - 406ms/epoch - 11ms/step
Epoch 20/20
36/36 - 0s - loss: 0.1900 - accuracy: 0.9178 - val_loss: 0.1644 - val_accuracy: 0.9553 - 321ms/epoch - 9ms/step
Validation accuracy for Model of [Learning Rate 0.1 | Num Filters 32 | Dropout Rate 0.7 | FilterSize 3]: 0.9552845358848572
Epoch 1/20
36/36 - 1s - loss: 1.1387 - accuracy: 0.7578 - val_loss: 0.2500 - val_accuracy: 0.9360 - 1s/epoch - 36ms/step
Epoch 2/20
36/36 - 0s - loss: 0.2382 - accuracy: 0.9060 - val_loss: 0.1947 - val_accuracy: 0.9436 - 366ms/epoch - 10ms/step
Epoch 3/20
36/36 - 0s - loss: 0.2163 - accuracy: 0.9126 - val_loss: 0.1756 - val_accuracy: 0.9548 - 379ms/epoch - 11ms/step
Epoch 4/20
36/36 - 0s - loss: 0.2066 - accuracy: 0.9143 - val_loss: 0.1884 - val_accuracy: 0.9487 - 372ms/epoch - 10ms/step
Epoch 5/20
36/36 - 0s - loss: 0.2231 - accuracy: 0.9129 - val_loss: 0.1841 - val_accuracy: 0.9482 - 368ms/epoch - 10ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1910 - accuracy: 0.9190 - val_loss: 0.1694 - val_accuracy: 0.9512 - 326ms/epoch - 9ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1944 - accuracy: 0.9204 - val_loss: 0.1714 - val_accuracy: 0.9492 - 371ms/epoch - 10ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1928 - accuracy: 0.9180 - val_loss: 0.2279 - val_accuracy: 0.9456 - 339ms/epoch - 9ms/step
Epoch 9/20
36/36 - 0s - loss: 0.2021 - accuracy: 0.9138 - val_loss: 0.1751 - val_accuracy: 0.9472 - 277ms/epoch - 8ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1936 - accuracy: 0.9194 - val_loss: 0.1668 - val_accuracy: 0.9583 - 274ms/epoch - 8ms/step
Epoch 11/20
36/36 - 0s - loss: 0.1919 - accuracy: 0.9214 - val_loss: 0.1755 - val_accuracy: 0.9563 - 290ms/epoch - 8ms/step
Epoch 12/20
36/36 - 0s - loss: 0.2016 - accuracy: 0.9156 - val_loss: 0.1712 - val_accuracy: 0.9538 - 293ms/epoch - 8ms/step
Epoch 13/20
36/36 - 0s - loss: 0.2040 - accuracy: 0.9140 - val_loss: 0.1447 - val_accuracy: 0.9553 - 279ms/epoch - 8ms/step
Epoch 14/20
36/36 - 0s - loss: 0.1919 - accuracy: 0.9174 - val_loss: 0.1236 - val_accuracy: 0.9644 - 288ms/epoch - 8ms/step
Epoch 15/20
36/36 - 0s - loss: 0.2020 - accuracy: 0.9073 - val_loss: 0.1877 - val_accuracy: 0.9512 - 294ms/epoch - 8ms/step
Epoch 16/20
36/36 - 0s - loss: 0.2132 - accuracy: 0.9089 - val_loss: 0.1752 - val_accuracy: 0.9568 - 273ms/epoch - 8ms/step
Epoch 17/20
36/36 - 0s - loss: 0.2127 - accuracy: 0.9041 - val_loss: 0.1573 - val_accuracy: 0.9548 - 315ms/epoch - 9ms/step
Epoch 18/20
36/36 - 0s - loss: 0.2492 - accuracy: 0.8840 - val_loss: 0.1900 - val_accuracy: 0.9497 - 277ms/epoch - 8ms/step
Epoch 19/20
36/36 - 0s - loss: 0.2217 - accuracy: 0.8974 - val_loss: 0.1516 - val_accuracy: 0.9512 - 332ms/epoch - 9ms/step
Epoch 20/20
36/36 - 0s - loss: 0.2095 - accuracy: 0.9016 - val_loss: 0.1543 - val_accuracy: 0.9563 - 418ms/epoch - 12ms/step
Validation accuracy for Model of [Learning Rate 0.1 | Num Filters 32 | Dropout Rate 0.7 | FilterSize 5]: 0.9563007950782776
Epoch 1/20
36/36 - 1s - loss: 0.5238 - accuracy: 0.8103 - val_loss: 0.2157 - val_accuracy: 0.9370 - 1s/epoch - 40ms/step
Epoch 2/20
36/36 - 0s - loss: 0.2472 - accuracy: 0.8972 - val_loss: 0.1684 - val_accuracy: 0.9466 - 385ms/epoch - 11ms/step
Epoch 3/20
36/36 - 0s - loss: 0.2168 - accuracy: 0.9030 - val_loss: 0.1769 - val_accuracy: 0.9492 - 334ms/epoch - 9ms/step
Epoch 4/20
36/36 - 0s - loss: 0.2009 - accuracy: 0.9188 - val_loss: 0.1537 - val_accuracy: 0.9543 - 358ms/epoch - 10ms/step
Epoch 5/20
36/36 - 0s - loss: 0.2017 - accuracy: 0.9151 - val_loss: 0.1791 - val_accuracy: 0.9573 - 297ms/epoch - 8ms/step
Epoch 6/20
36/36 - 0s - loss: 0.2236 - accuracy: 0.9080 - val_loss: 0.1928 - val_accuracy: 0.9405 - 283ms/epoch - 8ms/step
Epoch 7/20
36/36 - 0s - loss: 0.2394 - accuracy: 0.8977 - val_loss: 0.1392 - val_accuracy: 0.9507 - 325ms/epoch - 9ms/step
Epoch 8/20
36/36 - 0s - loss: 0.2637 - accuracy: 0.8861 - val_loss: 0.1932 - val_accuracy: 0.9512 - 282ms/epoch - 8ms/step
Epoch 9/20
36/36 - 0s - loss: 0.3381 - accuracy: 0.8071 - val_loss: 0.2583 - val_accuracy: 0.9370 - 283ms/epoch - 8ms/step
Epoch 10/20
36/36 - 0s - loss: 0.3552 - accuracy: 0.8180 - val_loss: 0.2572 - val_accuracy: 0.9431 - 271ms/epoch - 8ms/step
Epoch 11/20
36/36 - 0s - loss: 0.3085 - accuracy: 0.8230 - val_loss: 0.2316 - val_accuracy: 0.9502 - 273ms/epoch - 8ms/step
Epoch 12/20
36/36 - 0s - loss: 0.3016 - accuracy: 0.8382 - val_loss: 0.2284 - val_accuracy: 0.9527 - 311ms/epoch - 9ms/step
Epoch 13/20
36/36 - 0s - loss: 0.3236 - accuracy: 0.8522 - val_loss: 0.3028 - val_accuracy: 0.9482 - 383ms/epoch - 11ms/step
Epoch 14/20
36/36 - 0s - loss: 0.3541 - accuracy: 0.8073 - val_loss: 0.2328 - val_accuracy: 0.7683 - 322ms/epoch - 9ms/step
Epoch 15/20
36/36 - 0s - loss: 0.3220 - accuracy: 0.8007 - val_loss: 0.2738 - val_accuracy: 0.7556 - 364ms/epoch - 10ms/step
Epoch 16/20
36/36 - 0s - loss: 0.2994 - accuracy: 0.8102 - val_loss: 0.2220 - val_accuracy: 0.9533 - 327ms/epoch - 9ms/step
Epoch 17/20
36/36 - 0s - loss: 0.3246 - accuracy: 0.7970 - val_loss: 0.2549 - val_accuracy: 0.7713 - 329ms/epoch - 9ms/step
Epoch 18/20
36/36 - 0s - loss: 0.3041 - accuracy: 0.7997 - val_loss: 0.2588 - val_accuracy: 0.8023 - 332ms/epoch - 9ms/step
Epoch 19/20
36/36 - 0s - loss: 0.3381 - accuracy: 0.8029 - val_loss: 0.4064 - val_accuracy: 0.7754 - 376ms/epoch - 10ms/step
Epoch 20/20
36/36 - 0s - loss: 0.3495 - accuracy: 0.8045 - val_loss: 0.3124 - val_accuracy: 0.7109 - 357ms/epoch - 10ms/step
Validation accuracy for Model of [Learning Rate 0.1 | Num Filters 32 | Dropout Rate 0.7 | FilterSize 7]: 0.7108739614486694
Epoch 1/20
36/36 - 2s - loss: 1.0119 - accuracy: 0.7992 - val_loss: 0.1688 - val_accuracy: 0.9431 - 2s/epoch - 45ms/step
Epoch 2/20
36/36 - 0s - loss: 0.1716 - accuracy: 0.9310 - val_loss: 0.1423 - val_accuracy: 0.9512 - 446ms/epoch - 12ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1453 - accuracy: 0.9403 - val_loss: 0.1511 - val_accuracy: 0.9441 - 421ms/epoch - 12ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1405 - accuracy: 0.9419 - val_loss: 0.1302 - val_accuracy: 0.9426 - 420ms/epoch - 12ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1265 - accuracy: 0.9452 - val_loss: 0.1115 - val_accuracy: 0.9614 - 433ms/epoch - 12ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1262 - accuracy: 0.9486 - val_loss: 0.1110 - val_accuracy: 0.9548 - 431ms/epoch - 12ms/step
Epoch 7/20
36/36 - 1s - loss: 0.1208 - accuracy: 0.9460 - val_loss: 0.1183 - val_accuracy: 0.9487 - 502ms/epoch - 14ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1224 - accuracy: 0.9480 - val_loss: 0.1325 - val_accuracy: 0.9441 - 484ms/epoch - 13ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1257 - accuracy: 0.9488 - val_loss: 0.1299 - val_accuracy: 0.9527 - 468ms/epoch - 13ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1305 - accuracy: 0.9466 - val_loss: 0.1218 - val_accuracy: 0.9563 - 460ms/epoch - 13ms/step
Epoch 11/20
36/36 - 0s - loss: 0.1154 - accuracy: 0.9525 - val_loss: 0.1029 - val_accuracy: 0.9675 - 494ms/epoch - 14ms/step
Epoch 12/20
36/36 - 0s - loss: 0.1063 - accuracy: 0.9528 - val_loss: 0.1125 - val_accuracy: 0.9619 - 454ms/epoch - 13ms/step
Epoch 13/20
36/36 - 0s - loss: 0.1121 - accuracy: 0.9535 - val_loss: 0.1168 - val_accuracy: 0.9522 - 483ms/epoch - 13ms/step
Epoch 14/20
36/36 - 0s - loss: 0.1274 - accuracy: 0.9527 - val_loss: 0.1152 - val_accuracy: 0.9599 - 450ms/epoch - 12ms/step
Epoch 15/20
36/36 - 0s - loss: 0.1233 - accuracy: 0.9458 - val_loss: 0.1137 - val_accuracy: 0.9588 - 488ms/epoch - 14ms/step
Epoch 16/20
36/36 - 0s - loss: 0.1107 - accuracy: 0.9502 - val_loss: 0.1210 - val_accuracy: 0.9609 - 483ms/epoch - 13ms/step
Epoch 17/20
36/36 - 0s - loss: 0.1116 - accuracy: 0.9529 - val_loss: 0.1025 - val_accuracy: 0.9634 - 441ms/epoch - 12ms/step
Epoch 18/20
36/36 - 0s - loss: 0.1034 - accuracy: 0.9537 - val_loss: 0.1061 - val_accuracy: 0.9710 - 445ms/epoch - 12ms/step
Epoch 19/20
36/36 - 0s - loss: 0.1088 - accuracy: 0.9559 - val_loss: 0.1176 - val_accuracy: 0.9548 - 424ms/epoch - 12ms/step
Epoch 20/20
36/36 - 0s - loss: 0.1106 - accuracy: 0.9528 - val_loss: 0.1136 - val_accuracy: 0.9588 - 496ms/epoch - 14ms/step
Validation accuracy for Model of [Learning Rate 0.1 | Num Filters 64 | Dropout Rate 0.3 | FilterSize 3]: 0.9588414430618286
Epoch 1/20
36/36 - 2s - loss: 1.3186 - accuracy: 0.7987 - val_loss: 0.1714 - val_accuracy: 0.9365 - 2s/epoch - 56ms/step
Epoch 2/20
36/36 - 1s - loss: 0.1657 - accuracy: 0.9316 - val_loss: 0.1319 - val_accuracy: 0.9482 - 627ms/epoch - 17ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1450 - accuracy: 0.9362 - val_loss: 0.1134 - val_accuracy: 0.9538 - 698ms/epoch - 19ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1361 - accuracy: 0.9409 - val_loss: 0.1110 - val_accuracy: 0.9604 - 427ms/epoch - 12ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1254 - accuracy: 0.9483 - val_loss: 0.1668 - val_accuracy: 0.9456 - 486ms/epoch - 13ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1346 - accuracy: 0.9481 - val_loss: 0.1087 - val_accuracy: 0.9583 - 460ms/epoch - 13ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1330 - accuracy: 0.9434 - val_loss: 0.1417 - val_accuracy: 0.9472 - 423ms/epoch - 12ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1203 - accuracy: 0.9439 - val_loss: 0.1477 - val_accuracy: 0.9507 - 397ms/epoch - 11ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1289 - accuracy: 0.9399 - val_loss: 0.1545 - val_accuracy: 0.9492 - 382ms/epoch - 11ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1168 - accuracy: 0.9452 - val_loss: 0.1023 - val_accuracy: 0.9634 - 398ms/epoch - 11ms/step
Epoch 11/20
36/36 - 0s - loss: 0.1100 - accuracy: 0.9464 - val_loss: 0.1206 - val_accuracy: 0.9563 - 372ms/epoch - 10ms/step
Epoch 12/20
36/36 - 0s - loss: 0.1065 - accuracy: 0.9507 - val_loss: 0.1056 - val_accuracy: 0.9639 - 430ms/epoch - 12ms/step
Epoch 13/20
36/36 - 0s - loss: 0.0997 - accuracy: 0.9534 - val_loss: 0.1393 - val_accuracy: 0.9543 - 373ms/epoch - 10ms/step
Epoch 14/20
36/36 - 0s - loss: 0.1067 - accuracy: 0.9508 - val_loss: 0.1078 - val_accuracy: 0.9619 - 440ms/epoch - 12ms/step
Epoch 15/20
36/36 - 0s - loss: 0.1031 - accuracy: 0.9497 - val_loss: 0.1242 - val_accuracy: 0.9563 - 489ms/epoch - 14ms/step
Epoch 16/20
36/36 - 0s - loss: 0.0975 - accuracy: 0.9516 - val_loss: 0.1168 - val_accuracy: 0.9441 - 473ms/epoch - 13ms/step
Epoch 17/20
36/36 - 0s - loss: 0.1023 - accuracy: 0.9506 - val_loss: 0.1185 - val_accuracy: 0.9624 - 450ms/epoch - 12ms/step
Epoch 18/20
36/36 - 0s - loss: 0.1102 - accuracy: 0.9449 - val_loss: 0.1285 - val_accuracy: 0.9395 - 414ms/epoch - 11ms/step
Epoch 19/20
36/36 - 0s - loss: 0.0996 - accuracy: 0.9497 - val_loss: 0.1154 - val_accuracy: 0.9461 - 382ms/epoch - 11ms/step
Epoch 20/20
36/36 - 0s - loss: 0.0926 - accuracy: 0.9563 - val_loss: 0.1300 - val_accuracy: 0.9238 - 408ms/epoch - 11ms/step
Validation accuracy for Model of [Learning Rate 0.1 | Num Filters 64 | Dropout Rate 0.3 | FilterSize 5]: 0.9237805008888245
Epoch 1/20
36/36 - 1s - loss: 1.5615 - accuracy: 0.7916 - val_loss: 0.1848 - val_accuracy: 0.9304 - 1s/epoch - 39ms/step
Epoch 2/20
36/36 - 0s - loss: 0.1890 - accuracy: 0.9236 - val_loss: 0.1504 - val_accuracy: 0.9345 - 418ms/epoch - 12ms/step
Epoch 3/20
36/36 - 0s - loss: 0.1498 - accuracy: 0.9321 - val_loss: 0.1388 - val_accuracy: 0.9461 - 397ms/epoch - 11ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1380 - accuracy: 0.9382 - val_loss: 0.1488 - val_accuracy: 0.9492 - 528ms/epoch - 15ms/step
Epoch 5/20
36/36 - 1s - loss: 0.1294 - accuracy: 0.9421 - val_loss: 0.1420 - val_accuracy: 0.9533 - 684ms/epoch - 19ms/step
Epoch 6/20
36/36 - 1s - loss: 0.1206 - accuracy: 0.9444 - val_loss: 0.1333 - val_accuracy: 0.9533 - 721ms/epoch - 20ms/step
Epoch 7/20
36/36 - 1s - loss: 0.1191 - accuracy: 0.9418 - val_loss: 0.1342 - val_accuracy: 0.9512 - 643ms/epoch - 18ms/step
Epoch 8/20
36/36 - 1s - loss: 0.1127 - accuracy: 0.9464 - val_loss: 0.1182 - val_accuracy: 0.9451 - 627ms/epoch - 17ms/step
Epoch 9/20
36/36 - 1s - loss: 0.1124 - accuracy: 0.9499 - val_loss: 0.1307 - val_accuracy: 0.9461 - 604ms/epoch - 17ms/step
Epoch 10/20
36/36 - 1s - loss: 0.1103 - accuracy: 0.9534 - val_loss: 0.1013 - val_accuracy: 0.9619 - 758ms/epoch - 21ms/step
Epoch 11/20
36/36 - 1s - loss: 0.1020 - accuracy: 0.9543 - val_loss: 0.1320 - val_accuracy: 0.9451 - 683ms/epoch - 19ms/step
Epoch 12/20
36/36 - 1s - loss: 0.1063 - accuracy: 0.9514 - val_loss: 0.1082 - val_accuracy: 0.9497 - 732ms/epoch - 20ms/step
Epoch 13/20
36/36 - 1s - loss: 0.1003 - accuracy: 0.9535 - val_loss: 0.1342 - val_accuracy: 0.9299 - 544ms/epoch - 15ms/step
Epoch 14/20
36/36 - 0s - loss: 0.1060 - accuracy: 0.9503 - val_loss: 0.1492 - val_accuracy: 0.9395 - 442ms/epoch - 12ms/step
Epoch 15/20
36/36 - 0s - loss: 0.1134 - accuracy: 0.9510 - val_loss: 0.1384 - val_accuracy: 0.9461 - 442ms/epoch - 12ms/step
Epoch 16/20
36/36 - 0s - loss: 0.1116 - accuracy: 0.9483 - val_loss: 0.1423 - val_accuracy: 0.9456 - 435ms/epoch - 12ms/step
Epoch 17/20
36/36 - 0s - loss: 0.0987 - accuracy: 0.9494 - val_loss: 0.1436 - val_accuracy: 0.9472 - 371ms/epoch - 10ms/step
Epoch 18/20
36/36 - 0s - loss: 0.1048 - accuracy: 0.9473 - val_loss: 0.1263 - val_accuracy: 0.9426 - 337ms/epoch - 9ms/step
Epoch 19/20
36/36 - 0s - loss: 0.1006 - accuracy: 0.9521 - val_loss: 0.1214 - val_accuracy: 0.9385 - 394ms/epoch - 11ms/step
Epoch 20/20
36/36 - 0s - loss: 0.1042 - accuracy: 0.9492 - val_loss: 0.1277 - val_accuracy: 0.9466 - 447ms/epoch - 12ms/step
Validation accuracy for Model of [Learning Rate 0.1 | Num Filters 64 | Dropout Rate 0.3 | FilterSize 7]: 0.9466463327407837
Epoch 1/20
36/36 - 1s - loss: 0.9490 - accuracy: 0.8128 - val_loss: 0.1704 - val_accuracy: 0.9416 - 1s/epoch - 39ms/step
Epoch 2/20
36/36 - 0s - loss: 0.1861 - accuracy: 0.9228 - val_loss: 0.1461 - val_accuracy: 0.9395 - 462ms/epoch - 13ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1580 - accuracy: 0.9333 - val_loss: 0.1557 - val_accuracy: 0.9517 - 504ms/epoch - 14ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1594 - accuracy: 0.9344 - val_loss: 0.1328 - val_accuracy: 0.9568 - 490ms/epoch - 14ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1451 - accuracy: 0.9389 - val_loss: 0.1320 - val_accuracy: 0.9543 - 494ms/epoch - 14ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1445 - accuracy: 0.9386 - val_loss: 0.1356 - val_accuracy: 0.9578 - 482ms/epoch - 13ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1394 - accuracy: 0.9368 - val_loss: 0.1578 - val_accuracy: 0.9461 - 465ms/epoch - 13ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1426 - accuracy: 0.9383 - val_loss: 0.1143 - val_accuracy: 0.9548 - 488ms/epoch - 14ms/step
Epoch 9/20
36/36 - 1s - loss: 0.1343 - accuracy: 0.9396 - val_loss: 0.1133 - val_accuracy: 0.9644 - 522ms/epoch - 14ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1426 - accuracy: 0.9399 - val_loss: 0.1256 - val_accuracy: 0.9624 - 484ms/epoch - 13ms/step
Epoch 11/20
36/36 - 0s - loss: 0.1453 - accuracy: 0.9405 - val_loss: 0.1375 - val_accuracy: 0.9599 - 472ms/epoch - 13ms/step
Epoch 12/20
36/36 - 0s - loss: 0.1367 - accuracy: 0.9408 - val_loss: 0.1009 - val_accuracy: 0.9619 - 436ms/epoch - 12ms/step
Epoch 13/20
36/36 - 1s - loss: 0.1310 - accuracy: 0.9457 - val_loss: 0.1161 - val_accuracy: 0.9553 - 630ms/epoch - 17ms/step
Epoch 14/20
36/36 - 1s - loss: 0.1301 - accuracy: 0.9445 - val_loss: 0.1066 - val_accuracy: 0.9619 - 837ms/epoch - 23ms/step
Epoch 15/20
36/36 - 1s - loss: 0.1442 - accuracy: 0.9406 - val_loss: 0.1307 - val_accuracy: 0.9573 - 734ms/epoch - 20ms/step
Epoch 16/20
36/36 - 1s - loss: 0.1384 - accuracy: 0.9391 - val_loss: 0.1296 - val_accuracy: 0.9563 - 670ms/epoch - 19ms/step
Epoch 17/20
36/36 - 1s - loss: 0.1346 - accuracy: 0.9401 - val_loss: 0.1041 - val_accuracy: 0.9654 - 644ms/epoch - 18ms/step
Epoch 18/20
36/36 - 1s - loss: 0.1473 - accuracy: 0.9361 - val_loss: 0.1250 - val_accuracy: 0.9593 - 708ms/epoch - 20ms/step
Epoch 19/20
36/36 - 1s - loss: 0.1373 - accuracy: 0.9410 - val_loss: 0.1173 - val_accuracy: 0.9690 - 756ms/epoch - 21ms/step
Epoch 20/20
36/36 - 1s - loss: 0.1397 - accuracy: 0.9415 - val_loss: 0.1163 - val_accuracy: 0.9675 - 561ms/epoch - 16ms/step
Validation accuracy for Model of [Learning Rate 0.1 | Num Filters 64 | Dropout Rate 0.5 | FilterSize 3]: 0.9674796462059021
Epoch 1/20
36/36 - 2s - loss: 1.5163 - accuracy: 0.7174 - val_loss: 0.2448 - val_accuracy: 0.8811 - 2s/epoch - 51ms/step
Epoch 2/20
36/36 - 1s - loss: 0.2152 - accuracy: 0.9113 - val_loss: 0.1467 - val_accuracy: 0.9538 - 669ms/epoch - 19ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1637 - accuracy: 0.9294 - val_loss: 0.1344 - val_accuracy: 0.9472 - 600ms/epoch - 17ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1488 - accuracy: 0.9393 - val_loss: 0.1424 - val_accuracy: 0.9507 - 585ms/epoch - 16ms/step
Epoch 5/20
36/36 - 1s - loss: 0.1517 - accuracy: 0.9393 - val_loss: 0.1251 - val_accuracy: 0.9583 - 653ms/epoch - 18ms/step
Epoch 6/20
36/36 - 1s - loss: 0.1407 - accuracy: 0.9410 - val_loss: 0.1078 - val_accuracy: 0.9609 - 768ms/epoch - 21ms/step
Epoch 7/20
36/36 - 1s - loss: 0.1433 - accuracy: 0.9442 - val_loss: 0.1334 - val_accuracy: 0.9573 - 848ms/epoch - 24ms/step
Epoch 8/20
36/36 - 1s - loss: 0.1432 - accuracy: 0.9390 - val_loss: 0.1251 - val_accuracy: 0.9578 - 590ms/epoch - 16ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1418 - accuracy: 0.9387 - val_loss: 0.1745 - val_accuracy: 0.9502 - 488ms/epoch - 14ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1322 - accuracy: 0.9432 - val_loss: 0.1123 - val_accuracy: 0.9609 - 447ms/epoch - 12ms/step
Epoch 11/20
36/36 - 0s - loss: 0.1309 - accuracy: 0.9430 - val_loss: 0.1246 - val_accuracy: 0.9527 - 471ms/epoch - 13ms/step
Epoch 12/20
36/36 - 0s - loss: 0.1309 - accuracy: 0.9432 - val_loss: 0.1129 - val_accuracy: 0.9624 - 398ms/epoch - 11ms/step
Epoch 13/20
36/36 - 0s - loss: 0.1270 - accuracy: 0.9463 - val_loss: 0.1151 - val_accuracy: 0.9660 - 492ms/epoch - 14ms/step
Epoch 14/20
36/36 - 1s - loss: 0.1194 - accuracy: 0.9471 - val_loss: 0.1194 - val_accuracy: 0.9649 - 517ms/epoch - 14ms/step
Epoch 15/20
36/36 - 1s - loss: 0.1310 - accuracy: 0.9479 - val_loss: 0.1115 - val_accuracy: 0.9624 - 511ms/epoch - 14ms/step
Epoch 16/20
36/36 - 0s - loss: 0.1264 - accuracy: 0.9477 - val_loss: 0.1135 - val_accuracy: 0.9649 - 493ms/epoch - 14ms/step
Epoch 17/20
36/36 - 1s - loss: 0.1284 - accuracy: 0.9430 - val_loss: 0.1147 - val_accuracy: 0.9624 - 513ms/epoch - 14ms/step
Epoch 18/20
36/36 - 0s - loss: 0.1203 - accuracy: 0.9456 - val_loss: 0.1309 - val_accuracy: 0.9538 - 498ms/epoch - 14ms/step
Epoch 19/20
36/36 - 1s - loss: 0.1213 - accuracy: 0.9442 - val_loss: 0.1105 - val_accuracy: 0.9604 - 507ms/epoch - 14ms/step
Epoch 20/20
36/36 - 1s - loss: 0.1202 - accuracy: 0.9476 - val_loss: 0.1047 - val_accuracy: 0.9675 - 512ms/epoch - 14ms/step
Validation accuracy for Model of [Learning Rate 0.1 | Num Filters 64 | Dropout Rate 0.5 | FilterSize 5]: 0.9674796462059021
Epoch 1/20
36/36 - 2s - loss: 0.8517 - accuracy: 0.7265 - val_loss: 0.1746 - val_accuracy: 0.9390 - 2s/epoch - 45ms/step
Epoch 2/20
36/36 - 0s - loss: 0.2560 - accuracy: 0.8681 - val_loss: 0.1777 - val_accuracy: 0.9243 - 424ms/epoch - 12ms/step
Epoch 3/20
36/36 - 0s - loss: 0.2163 - accuracy: 0.8835 - val_loss: 0.1702 - val_accuracy: 0.9441 - 449ms/epoch - 12ms/step
Epoch 4/20
36/36 - 0s - loss: 0.1906 - accuracy: 0.9068 - val_loss: 0.1295 - val_accuracy: 0.9568 - 432ms/epoch - 12ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1821 - accuracy: 0.9127 - val_loss: 0.1159 - val_accuracy: 0.9553 - 376ms/epoch - 10ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1842 - accuracy: 0.9096 - val_loss: 0.1215 - val_accuracy: 0.9543 - 433ms/epoch - 12ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1562 - accuracy: 0.9226 - val_loss: 0.1099 - val_accuracy: 0.9568 - 399ms/epoch - 11ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1553 - accuracy: 0.9242 - val_loss: 0.1673 - val_accuracy: 0.9548 - 415ms/epoch - 12ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1708 - accuracy: 0.9231 - val_loss: 0.1173 - val_accuracy: 0.9593 - 391ms/epoch - 11ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1660 - accuracy: 0.9197 - val_loss: 0.1240 - val_accuracy: 0.9507 - 418ms/epoch - 12ms/step
Epoch 11/20
36/36 - 0s - loss: 0.1665 - accuracy: 0.9163 - val_loss: 0.1377 - val_accuracy: 0.9558 - 400ms/epoch - 11ms/step
Epoch 12/20
36/36 - 0s - loss: 0.1749 - accuracy: 0.9067 - val_loss: 0.1528 - val_accuracy: 0.9558 - 409ms/epoch - 11ms/step
Epoch 13/20
36/36 - 0s - loss: 0.1626 - accuracy: 0.9157 - val_loss: 0.1176 - val_accuracy: 0.9644 - 400ms/epoch - 11ms/step
Epoch 14/20
36/36 - 0s - loss: 0.1541 - accuracy: 0.9193 - val_loss: 0.1317 - val_accuracy: 0.9177 - 412ms/epoch - 11ms/step
Epoch 15/20
36/36 - 0s - loss: 0.1491 - accuracy: 0.9211 - val_loss: 0.1495 - val_accuracy: 0.9096 - 455ms/epoch - 13ms/step
Epoch 16/20
36/36 - 1s - loss: 0.1452 - accuracy: 0.9240 - val_loss: 0.1808 - val_accuracy: 0.9278 - 640ms/epoch - 18ms/step
Epoch 17/20
36/36 - 1s - loss: 0.1863 - accuracy: 0.9045 - val_loss: 0.1649 - val_accuracy: 0.9446 - 624ms/epoch - 17ms/step
Epoch 18/20
36/36 - 1s - loss: 0.1926 - accuracy: 0.9010 - val_loss: 0.1437 - val_accuracy: 0.9588 - 619ms/epoch - 17ms/step
Epoch 19/20
36/36 - 1s - loss: 0.2003 - accuracy: 0.8996 - val_loss: 0.2111 - val_accuracy: 0.9614 - 613ms/epoch - 17ms/step
Epoch 20/20
36/36 - 1s - loss: 0.2279 - accuracy: 0.9038 - val_loss: 0.2369 - val_accuracy: 0.9477 - 625ms/epoch - 17ms/step
Validation accuracy for Model of [Learning Rate 0.1 | Num Filters 64 | Dropout Rate 0.5 | FilterSize 7]: 0.9476625919342041
Epoch 1/20
36/36 - 2s - loss: 0.6447 - accuracy: 0.7894 - val_loss: 0.2187 - val_accuracy: 0.9309 - 2s/epoch - 51ms/step
Epoch 2/20
36/36 - 0s - loss: 0.2417 - accuracy: 0.9088 - val_loss: 0.1712 - val_accuracy: 0.9411 - 384ms/epoch - 11ms/step
Epoch 3/20
36/36 - 0s - loss: 0.2193 - accuracy: 0.9087 - val_loss: 0.1443 - val_accuracy: 0.9538 - 390ms/epoch - 11ms/step
Epoch 4/20
36/36 - 0s - loss: 0.2039 - accuracy: 0.9208 - val_loss: 0.1751 - val_accuracy: 0.9395 - 419ms/epoch - 12ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1864 - accuracy: 0.9224 - val_loss: 0.1556 - val_accuracy: 0.9543 - 377ms/epoch - 10ms/step
Epoch 6/20
36/36 - 0s - loss: 0.2038 - accuracy: 0.9179 - val_loss: 0.1492 - val_accuracy: 0.9558 - 448ms/epoch - 12ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1876 - accuracy: 0.9179 - val_loss: 0.1508 - val_accuracy: 0.9568 - 380ms/epoch - 11ms/step
Epoch 8/20
36/36 - 0s - loss: 0.2010 - accuracy: 0.9169 - val_loss: 0.1648 - val_accuracy: 0.9472 - 381ms/epoch - 11ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1879 - accuracy: 0.9200 - val_loss: 0.1507 - val_accuracy: 0.9563 - 416ms/epoch - 12ms/step
Epoch 10/20
36/36 - 0s - loss: 0.2069 - accuracy: 0.9182 - val_loss: 0.1730 - val_accuracy: 0.9527 - 412ms/epoch - 11ms/step
Epoch 11/20
36/36 - 1s - loss: 0.2303 - accuracy: 0.9021 - val_loss: 0.2013 - val_accuracy: 0.9451 - 508ms/epoch - 14ms/step
Epoch 12/20
36/36 - 0s - loss: 0.2441 - accuracy: 0.8993 - val_loss: 0.1758 - val_accuracy: 0.9593 - 461ms/epoch - 13ms/step
Epoch 13/20
36/36 - 0s - loss: 0.2169 - accuracy: 0.9046 - val_loss: 0.1996 - val_accuracy: 0.9507 - 436ms/epoch - 12ms/step
Epoch 14/20
36/36 - 0s - loss: 0.2226 - accuracy: 0.9043 - val_loss: 0.1523 - val_accuracy: 0.9553 - 453ms/epoch - 13ms/step
Epoch 15/20
36/36 - 0s - loss: 0.2143 - accuracy: 0.9059 - val_loss: 0.1431 - val_accuracy: 0.9533 - 429ms/epoch - 12ms/step
Epoch 16/20
36/36 - 0s - loss: 0.2061 - accuracy: 0.9074 - val_loss: 0.1512 - val_accuracy: 0.9533 - 468ms/epoch - 13ms/step
Epoch 17/20
36/36 - 0s - loss: 0.1885 - accuracy: 0.9127 - val_loss: 0.1592 - val_accuracy: 0.9507 - 465ms/epoch - 13ms/step
Epoch 18/20
36/36 - 1s - loss: 0.1860 - accuracy: 0.9117 - val_loss: 0.1889 - val_accuracy: 0.9446 - 513ms/epoch - 14ms/step
Epoch 19/20
36/36 - 0s - loss: 0.1942 - accuracy: 0.9115 - val_loss: 0.1347 - val_accuracy: 0.9604 - 461ms/epoch - 13ms/step
Epoch 20/20
36/36 - 0s - loss: 0.1936 - accuracy: 0.9122 - val_loss: 0.1524 - val_accuracy: 0.9583 - 448ms/epoch - 12ms/step
Validation accuracy for Model of [Learning Rate 0.1 | Num Filters 64 | Dropout Rate 0.7 | FilterSize 3]: 0.9583333134651184
Epoch 1/20
36/36 - 2s - loss: 1.2395 - accuracy: 0.7521 - val_loss: 0.2411 - val_accuracy: 0.9339 - 2s/epoch - 60ms/step
Epoch 2/20
36/36 - 1s - loss: 0.2265 - accuracy: 0.9064 - val_loss: 0.1572 - val_accuracy: 0.9426 - 584ms/epoch - 16ms/step
Epoch 3/20
36/36 - 1s - loss: 0.2050 - accuracy: 0.9155 - val_loss: 0.1756 - val_accuracy: 0.9426 - 586ms/epoch - 16ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1944 - accuracy: 0.9220 - val_loss: 0.1263 - val_accuracy: 0.9538 - 592ms/epoch - 16ms/step
Epoch 5/20
36/36 - 1s - loss: 0.1956 - accuracy: 0.9177 - val_loss: 0.1747 - val_accuracy: 0.9533 - 506ms/epoch - 14ms/step
Epoch 6/20
36/36 - 0s - loss: 0.1892 - accuracy: 0.9222 - val_loss: 0.1375 - val_accuracy: 0.9583 - 418ms/epoch - 12ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1819 - accuracy: 0.9232 - val_loss: 0.1273 - val_accuracy: 0.9578 - 434ms/epoch - 12ms/step
Epoch 8/20
36/36 - 0s - loss: 0.1815 - accuracy: 0.9255 - val_loss: 0.1545 - val_accuracy: 0.9548 - 453ms/epoch - 13ms/step
Epoch 9/20
36/36 - 0s - loss: 0.1853 - accuracy: 0.9226 - val_loss: 0.1497 - val_accuracy: 0.9558 - 374ms/epoch - 10ms/step
Epoch 10/20
36/36 - 0s - loss: 0.1714 - accuracy: 0.9268 - val_loss: 0.1397 - val_accuracy: 0.9578 - 433ms/epoch - 12ms/step
Epoch 11/20
36/36 - 0s - loss: 0.1669 - accuracy: 0.9295 - val_loss: 0.1939 - val_accuracy: 0.9492 - 439ms/epoch - 12ms/step
Epoch 12/20
36/36 - 0s - loss: 0.1696 - accuracy: 0.9279 - val_loss: 0.1331 - val_accuracy: 0.9533 - 431ms/epoch - 12ms/step
Epoch 13/20
36/36 - 0s - loss: 0.1794 - accuracy: 0.9216 - val_loss: 0.1590 - val_accuracy: 0.9492 - 434ms/epoch - 12ms/step
Epoch 14/20
36/36 - 0s - loss: 0.1811 - accuracy: 0.9262 - val_loss: 0.1592 - val_accuracy: 0.9563 - 436ms/epoch - 12ms/step
Epoch 15/20
36/36 - 0s - loss: 0.2277 - accuracy: 0.9161 - val_loss: 0.1575 - val_accuracy: 0.9568 - 410ms/epoch - 11ms/step
Epoch 16/20
36/36 - 0s - loss: 0.2121 - accuracy: 0.9181 - val_loss: 0.1863 - val_accuracy: 0.9538 - 439ms/epoch - 12ms/step
Epoch 17/20
36/36 - 0s - loss: 0.1939 - accuracy: 0.9141 - val_loss: 0.1577 - val_accuracy: 0.9533 - 381ms/epoch - 11ms/step
Epoch 18/20
36/36 - 0s - loss: 0.1901 - accuracy: 0.9202 - val_loss: 0.1579 - val_accuracy: 0.9583 - 418ms/epoch - 12ms/step
Epoch 19/20
36/36 - 0s - loss: 0.2056 - accuracy: 0.9128 - val_loss: 0.1467 - val_accuracy: 0.9543 - 380ms/epoch - 11ms/step
Epoch 20/20
36/36 - 0s - loss: 0.2018 - accuracy: 0.9057 - val_loss: 0.1919 - val_accuracy: 0.9512 - 405ms/epoch - 11ms/step
Validation accuracy for Model of [Learning Rate 0.1 | Num Filters 64 | Dropout Rate 0.7 | FilterSize 5]: 0.9512194991111755
Epoch 1/20
36/36 - 1s - loss: 0.6803 - accuracy: 0.7899 - val_loss: 0.1954 - val_accuracy: 0.9522 - 1s/epoch - 37ms/step
Epoch 2/20
36/36 - 0s - loss: 0.2362 - accuracy: 0.9074 - val_loss: 0.1833 - val_accuracy: 0.9436 - 364ms/epoch - 10ms/step
Epoch 3/20
36/36 - 0s - loss: 0.2160 - accuracy: 0.9164 - val_loss: 0.1445 - val_accuracy: 0.9538 - 401ms/epoch - 11ms/step
Epoch 4/20
36/36 - 0s - loss: 0.2185 - accuracy: 0.9045 - val_loss: 0.1511 - val_accuracy: 0.9527 - 360ms/epoch - 10ms/step
Epoch 5/20
36/36 - 0s - loss: 0.2001 - accuracy: 0.9185 - val_loss: 0.1471 - val_accuracy: 0.9553 - 350ms/epoch - 10ms/step
Epoch 6/20
36/36 - 0s - loss: 0.2103 - accuracy: 0.9142 - val_loss: 0.1698 - val_accuracy: 0.9558 - 405ms/epoch - 11ms/step
Epoch 7/20
36/36 - 1s - loss: 0.2052 - accuracy: 0.9180 - val_loss: 0.1714 - val_accuracy: 0.9578 - 513ms/epoch - 14ms/step
Epoch 8/20
36/36 - 1s - loss: 0.2102 - accuracy: 0.9061 - val_loss: 0.1670 - val_accuracy: 0.9593 - 574ms/epoch - 16ms/step
Epoch 9/20
36/36 - 1s - loss: 0.1995 - accuracy: 0.9208 - val_loss: 0.1652 - val_accuracy: 0.9599 - 598ms/epoch - 17ms/step
Epoch 10/20
36/36 - 1s - loss: 0.2220 - accuracy: 0.9122 - val_loss: 0.1705 - val_accuracy: 0.9543 - 596ms/epoch - 17ms/step
Epoch 11/20
36/36 - 1s - loss: 0.1989 - accuracy: 0.9194 - val_loss: 0.1786 - val_accuracy: 0.9502 - 603ms/epoch - 17ms/step
Epoch 12/20
36/36 - 1s - loss: 0.2039 - accuracy: 0.9129 - val_loss: 0.2219 - val_accuracy: 0.9522 - 610ms/epoch - 17ms/step
Epoch 13/20
36/36 - 1s - loss: 0.2215 - accuracy: 0.9109 - val_loss: 0.1583 - val_accuracy: 0.9538 - 599ms/epoch - 17ms/step
Epoch 14/20
36/36 - 1s - loss: 0.2281 - accuracy: 0.9085 - val_loss: 0.2339 - val_accuracy: 0.9502 - 694ms/epoch - 19ms/step
Epoch 15/20
36/36 - 1s - loss: 0.3021 - accuracy: 0.8771 - val_loss: 0.2203 - val_accuracy: 0.9502 - 671ms/epoch - 19ms/step
Epoch 16/20
36/36 - 1s - loss: 0.2812 - accuracy: 0.8599 - val_loss: 0.2157 - val_accuracy: 0.9472 - 646ms/epoch - 18ms/step
Epoch 17/20
36/36 - 1s - loss: 0.2713 - accuracy: 0.8711 - val_loss: 0.2152 - val_accuracy: 0.9492 - 574ms/epoch - 16ms/step
Epoch 18/20
36/36 - 1s - loss: 0.2452 - accuracy: 0.8780 - val_loss: 0.3420 - val_accuracy: 0.9507 - 546ms/epoch - 15ms/step
Epoch 19/20
36/36 - 1s - loss: 0.4254 - accuracy: 0.8340 - val_loss: 0.3600 - val_accuracy: 0.8989 - 571ms/epoch - 16ms/step
Epoch 20/20
36/36 - 1s - loss: 0.4816 - accuracy: 0.7188 - val_loss: 0.3817 - val_accuracy: 0.8928 - 575ms/epoch - 16ms/step
Validation accuracy for Model of [Learning Rate 0.1 | Num Filters 64 | Dropout Rate 0.7 | FilterSize 7]: 0.8927845358848572
Epoch 1/20
36/36 - 3s - loss: 1.9836 - accuracy: 0.8075 - val_loss: 0.1632 - val_accuracy: 0.9416 - 3s/epoch - 93ms/step
Epoch 2/20
36/36 - 1s - loss: 0.1543 - accuracy: 0.9378 - val_loss: 0.1391 - val_accuracy: 0.9390 - 613ms/epoch - 17ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1440 - accuracy: 0.9408 - val_loss: 0.1428 - val_accuracy: 0.9456 - 604ms/epoch - 17ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1290 - accuracy: 0.9470 - val_loss: 0.1069 - val_accuracy: 0.9578 - 601ms/epoch - 17ms/step
Epoch 5/20
36/36 - 1s - loss: 0.1399 - accuracy: 0.9451 - val_loss: 0.1185 - val_accuracy: 0.9599 - 628ms/epoch - 17ms/step
Epoch 6/20
36/36 - 1s - loss: 0.1195 - accuracy: 0.9503 - val_loss: 0.1472 - val_accuracy: 0.9375 - 610ms/epoch - 17ms/step
Epoch 7/20
36/36 - 1s - loss: 0.1180 - accuracy: 0.9525 - val_loss: 0.1077 - val_accuracy: 0.9538 - 729ms/epoch - 20ms/step
Epoch 8/20
36/36 - 1s - loss: 0.1383 - accuracy: 0.9475 - val_loss: 0.1262 - val_accuracy: 0.9543 - 754ms/epoch - 21ms/step
Epoch 9/20
36/36 - 1s - loss: 0.1244 - accuracy: 0.9511 - val_loss: 0.0955 - val_accuracy: 0.9665 - 716ms/epoch - 20ms/step
Epoch 10/20
36/36 - 1s - loss: 0.1111 - accuracy: 0.9550 - val_loss: 0.1022 - val_accuracy: 0.9568 - 673ms/epoch - 19ms/step
Epoch 11/20
36/36 - 1s - loss: 0.1115 - accuracy: 0.9550 - val_loss: 0.1076 - val_accuracy: 0.9604 - 662ms/epoch - 18ms/step
Epoch 12/20
36/36 - 1s - loss: 0.1112 - accuracy: 0.9551 - val_loss: 0.1055 - val_accuracy: 0.9614 - 641ms/epoch - 18ms/step
Epoch 13/20
36/36 - 1s - loss: 0.0987 - accuracy: 0.9605 - val_loss: 0.1061 - val_accuracy: 0.9634 - 673ms/epoch - 19ms/step
Epoch 14/20
36/36 - 1s - loss: 0.1045 - accuracy: 0.9579 - val_loss: 0.0925 - val_accuracy: 0.9736 - 918ms/epoch - 26ms/step
Epoch 15/20
36/36 - 1s - loss: 0.1053 - accuracy: 0.9555 - val_loss: 0.1021 - val_accuracy: 0.9619 - 1s/epoch - 28ms/step
Epoch 16/20
36/36 - 1s - loss: 0.1001 - accuracy: 0.9596 - val_loss: 0.0992 - val_accuracy: 0.9695 - 1s/epoch - 28ms/step
Epoch 17/20
36/36 - 1s - loss: 0.0994 - accuracy: 0.9616 - val_loss: 0.0937 - val_accuracy: 0.9695 - 1s/epoch - 28ms/step
Epoch 18/20
36/36 - 1s - loss: 0.0911 - accuracy: 0.9644 - val_loss: 0.1217 - val_accuracy: 0.9634 - 995ms/epoch - 28ms/step
Epoch 19/20
36/36 - 1s - loss: 0.0983 - accuracy: 0.9615 - val_loss: 0.1204 - val_accuracy: 0.9543 - 909ms/epoch - 25ms/step
Epoch 20/20
36/36 - 1s - loss: 0.0952 - accuracy: 0.9630 - val_loss: 0.0989 - val_accuracy: 0.9695 - 617ms/epoch - 17ms/step
Validation accuracy for Model of [Learning Rate 0.1 | Num Filters 128 | Dropout Rate 0.3 | FilterSize 3]: 0.9695122241973877
Epoch 1/20
36/36 - 2s - loss: 2.8663 - accuracy: 0.7484 - val_loss: 0.2822 - val_accuracy: 0.8435 - 2s/epoch - 45ms/step
Epoch 2/20
36/36 - 1s - loss: 0.2530 - accuracy: 0.8934 - val_loss: 0.1432 - val_accuracy: 0.9466 - 649ms/epoch - 18ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1839 - accuracy: 0.9146 - val_loss: 0.1217 - val_accuracy: 0.9553 - 627ms/epoch - 17ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1531 - accuracy: 0.9234 - val_loss: 0.1166 - val_accuracy: 0.9543 - 622ms/epoch - 17ms/step
Epoch 5/20
36/36 - 1s - loss: 0.1405 - accuracy: 0.9330 - val_loss: 0.1270 - val_accuracy: 0.9604 - 642ms/epoch - 18ms/step
Epoch 6/20
36/36 - 1s - loss: 0.1459 - accuracy: 0.9285 - val_loss: 0.1537 - val_accuracy: 0.9543 - 651ms/epoch - 18ms/step
Epoch 7/20
36/36 - 1s - loss: 0.1576 - accuracy: 0.9239 - val_loss: 0.1845 - val_accuracy: 0.9527 - 645ms/epoch - 18ms/step
Epoch 8/20
36/36 - 1s - loss: 0.1837 - accuracy: 0.9195 - val_loss: 0.1555 - val_accuracy: 0.9543 - 587ms/epoch - 16ms/step
Epoch 9/20
36/36 - 1s - loss: 0.1546 - accuracy: 0.9255 - val_loss: 0.1520 - val_accuracy: 0.9548 - 688ms/epoch - 19ms/step
Epoch 10/20
36/36 - 1s - loss: 0.1415 - accuracy: 0.9253 - val_loss: 0.1142 - val_accuracy: 0.9477 - 720ms/epoch - 20ms/step
Epoch 11/20
36/36 - 1s - loss: 0.1308 - accuracy: 0.9318 - val_loss: 0.1277 - val_accuracy: 0.9400 - 686ms/epoch - 19ms/step
Epoch 12/20
36/36 - 1s - loss: 0.1235 - accuracy: 0.9352 - val_loss: 0.1309 - val_accuracy: 0.9451 - 586ms/epoch - 16ms/step
Epoch 13/20
36/36 - 1s - loss: 0.1270 - accuracy: 0.9358 - val_loss: 0.1760 - val_accuracy: 0.9599 - 717ms/epoch - 20ms/step
Epoch 14/20
36/36 - 1s - loss: 0.1274 - accuracy: 0.9364 - val_loss: 0.1419 - val_accuracy: 0.9573 - 916ms/epoch - 25ms/step
Epoch 15/20
36/36 - 1s - loss: 0.1215 - accuracy: 0.9400 - val_loss: 0.1292 - val_accuracy: 0.9599 - 980ms/epoch - 27ms/step
Epoch 16/20
36/36 - 1s - loss: 0.1222 - accuracy: 0.9385 - val_loss: 0.1349 - val_accuracy: 0.9619 - 998ms/epoch - 28ms/step
Epoch 17/20
36/36 - 1s - loss: 0.1354 - accuracy: 0.9353 - val_loss: 0.1167 - val_accuracy: 0.9390 - 932ms/epoch - 26ms/step
Epoch 18/20
36/36 - 1s - loss: 0.1183 - accuracy: 0.9394 - val_loss: 0.1698 - val_accuracy: 0.8765 - 936ms/epoch - 26ms/step
Epoch 19/20
36/36 - 1s - loss: 0.1360 - accuracy: 0.9287 - val_loss: 0.1353 - val_accuracy: 0.9599 - 676ms/epoch - 19ms/step
Epoch 20/20
36/36 - 1s - loss: 0.1287 - accuracy: 0.9292 - val_loss: 0.1282 - val_accuracy: 0.9355 - 702ms/epoch - 20ms/step
Validation accuracy for Model of [Learning Rate 0.1 | Num Filters 128 | Dropout Rate 0.3 | FilterSize 5]: 0.9354674816131592
Epoch 1/20
36/36 - 2s - loss: 2.9478 - accuracy: 0.7396 - val_loss: 0.2351 - val_accuracy: 0.9126 - 2s/epoch - 52ms/step
Epoch 2/20
36/36 - 1s - loss: 0.2012 - accuracy: 0.9064 - val_loss: 0.1431 - val_accuracy: 0.9375 - 597ms/epoch - 17ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1544 - accuracy: 0.9274 - val_loss: 0.1362 - val_accuracy: 0.9365 - 611ms/epoch - 17ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1439 - accuracy: 0.9302 - val_loss: 0.1629 - val_accuracy: 0.9217 - 602ms/epoch - 17ms/step
Epoch 5/20
36/36 - 1s - loss: 0.1505 - accuracy: 0.9171 - val_loss: 0.1341 - val_accuracy: 0.9538 - 628ms/epoch - 17ms/step
Epoch 6/20
36/36 - 1s - loss: 0.1418 - accuracy: 0.9281 - val_loss: 0.1439 - val_accuracy: 0.9339 - 690ms/epoch - 19ms/step
Epoch 7/20
36/36 - 1s - loss: 0.1243 - accuracy: 0.9360 - val_loss: 0.1359 - val_accuracy: 0.9400 - 673ms/epoch - 19ms/step
Epoch 8/20
36/36 - 1s - loss: 0.1267 - accuracy: 0.9356 - val_loss: 0.1750 - val_accuracy: 0.8882 - 601ms/epoch - 17ms/step
Epoch 9/20
36/36 - 1s - loss: 0.1309 - accuracy: 0.9383 - val_loss: 0.1747 - val_accuracy: 0.9268 - 616ms/epoch - 17ms/step
Epoch 10/20
36/36 - 1s - loss: 0.1158 - accuracy: 0.9412 - val_loss: 0.1269 - val_accuracy: 0.9350 - 624ms/epoch - 17ms/step
Epoch 11/20
36/36 - 1s - loss: 0.1191 - accuracy: 0.9388 - val_loss: 0.1456 - val_accuracy: 0.9197 - 620ms/epoch - 17ms/step
Epoch 12/20
36/36 - 1s - loss: 0.1172 - accuracy: 0.9439 - val_loss: 0.1613 - val_accuracy: 0.9319 - 642ms/epoch - 18ms/step
Epoch 13/20
36/36 - 1s - loss: 0.1103 - accuracy: 0.9438 - val_loss: 0.1164 - val_accuracy: 0.9329 - 863ms/epoch - 24ms/step
Epoch 14/20
36/36 - 1s - loss: 0.1098 - accuracy: 0.9451 - val_loss: 0.1539 - val_accuracy: 0.9375 - 840ms/epoch - 23ms/step
Epoch 15/20
36/36 - 1s - loss: 0.1224 - accuracy: 0.9376 - val_loss: 0.1172 - val_accuracy: 0.9654 - 845ms/epoch - 23ms/step
Epoch 16/20
36/36 - 1s - loss: 0.1172 - accuracy: 0.9362 - val_loss: 0.1431 - val_accuracy: 0.9167 - 847ms/epoch - 24ms/step
Epoch 17/20
36/36 - 1s - loss: 0.1244 - accuracy: 0.9373 - val_loss: 0.1426 - val_accuracy: 0.9624 - 959ms/epoch - 27ms/step
Epoch 18/20
36/36 - 1s - loss: 0.1156 - accuracy: 0.9387 - val_loss: 0.1307 - val_accuracy: 0.9360 - 758ms/epoch - 21ms/step
Epoch 19/20
36/36 - 1s - loss: 0.1090 - accuracy: 0.9459 - val_loss: 0.1354 - val_accuracy: 0.9375 - 581ms/epoch - 16ms/step
Epoch 20/20
36/36 - 1s - loss: 0.1164 - accuracy: 0.9447 - val_loss: 0.1860 - val_accuracy: 0.8953 - 586ms/epoch - 16ms/step
Validation accuracy for Model of [Learning Rate 0.1 | Num Filters 128 | Dropout Rate 0.3 | FilterSize 7]: 0.8953251838684082
Epoch 1/20
36/36 - 2s - loss: 3.4459 - accuracy: 0.7478 - val_loss: 0.1817 - val_accuracy: 0.9268 - 2s/epoch - 52ms/step
Epoch 2/20
36/36 - 1s - loss: 0.2273 - accuracy: 0.9125 - val_loss: 0.1537 - val_accuracy: 0.9431 - 796ms/epoch - 22ms/step
Epoch 3/20
36/36 - 1s - loss: 0.1681 - accuracy: 0.9289 - val_loss: 0.1601 - val_accuracy: 0.9436 - 696ms/epoch - 19ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1598 - accuracy: 0.9340 - val_loss: 0.1513 - val_accuracy: 0.9461 - 704ms/epoch - 20ms/step
Epoch 5/20
36/36 - 1s - loss: 0.1542 - accuracy: 0.9339 - val_loss: 0.1510 - val_accuracy: 0.9456 - 684ms/epoch - 19ms/step
Epoch 6/20
36/36 - 1s - loss: 0.1575 - accuracy: 0.9340 - val_loss: 0.1611 - val_accuracy: 0.9395 - 715ms/epoch - 20ms/step
Epoch 7/20
36/36 - 1s - loss: 0.1534 - accuracy: 0.9331 - val_loss: 0.1303 - val_accuracy: 0.9461 - 733ms/epoch - 20ms/step
Epoch 8/20
36/36 - 1s - loss: 0.1399 - accuracy: 0.9373 - val_loss: 0.1246 - val_accuracy: 0.9568 - 703ms/epoch - 20ms/step
Epoch 9/20
36/36 - 1s - loss: 0.1425 - accuracy: 0.9329 - val_loss: 0.1228 - val_accuracy: 0.9482 - 685ms/epoch - 19ms/step
Epoch 10/20
36/36 - 1s - loss: 0.1284 - accuracy: 0.9377 - val_loss: 0.1634 - val_accuracy: 0.9451 - 681ms/epoch - 19ms/step
Epoch 11/20
36/36 - 1s - loss: 0.1318 - accuracy: 0.9370 - val_loss: 0.1091 - val_accuracy: 0.9517 - 966ms/epoch - 27ms/step
Epoch 12/20
36/36 - 1s - loss: 0.1341 - accuracy: 0.9359 - val_loss: 0.1228 - val_accuracy: 0.9502 - 1s/epoch - 28ms/step
Epoch 13/20
36/36 - 1s - loss: 0.1321 - accuracy: 0.9346 - val_loss: 0.1453 - val_accuracy: 0.9466 - 1s/epoch - 29ms/step
Epoch 14/20
36/36 - 1s - loss: 0.1341 - accuracy: 0.9400 - val_loss: 0.1275 - val_accuracy: 0.9472 - 1s/epoch - 29ms/step
Epoch 15/20
36/36 - 1s - loss: 0.1407 - accuracy: 0.9351 - val_loss: 0.1158 - val_accuracy: 0.9507 - 1s/epoch - 29ms/step
Epoch 16/20
36/36 - 1s - loss: 0.1360 - accuracy: 0.9351 - val_loss: 0.1445 - val_accuracy: 0.9472 - 964ms/epoch - 27ms/step
Epoch 17/20
36/36 - 1s - loss: 0.1438 - accuracy: 0.9332 - val_loss: 0.1229 - val_accuracy: 0.9578 - 679ms/epoch - 19ms/step
Epoch 18/20
36/36 - 1s - loss: 0.1267 - accuracy: 0.9365 - val_loss: 0.1199 - val_accuracy: 0.9497 - 636ms/epoch - 18ms/step
Epoch 19/20
36/36 - 1s - loss: 0.1306 - accuracy: 0.9375 - val_loss: 0.1131 - val_accuracy: 0.9512 - 775ms/epoch - 22ms/step
Epoch 20/20
36/36 - 1s - loss: 0.1265 - accuracy: 0.9369 - val_loss: 0.1250 - val_accuracy: 0.9360 - 663ms/epoch - 18ms/step
Validation accuracy for Model of [Learning Rate 0.1 | Num Filters 128 | Dropout Rate 0.5 | FilterSize 3]: 0.9359756112098694
Epoch 1/20
36/36 - 2s - loss: 1.7760 - accuracy: 0.7579 - val_loss: 0.2515 - val_accuracy: 0.9207 - 2s/epoch - 47ms/step
Epoch 2/20
36/36 - 1s - loss: 0.2362 - accuracy: 0.8926 - val_loss: 0.1872 - val_accuracy: 0.9304 - 665ms/epoch - 18ms/step
Epoch 3/20
36/36 - 1s - loss: 0.2075 - accuracy: 0.9118 - val_loss: 0.1549 - val_accuracy: 0.9507 - 764ms/epoch - 21ms/step
Epoch 4/20
36/36 - 1s - loss: 0.1833 - accuracy: 0.9224 - val_loss: 0.1477 - val_accuracy: 0.9538 - 737ms/epoch - 20ms/step
Epoch 5/20
36/36 - 1s - loss: 0.1755 - accuracy: 0.9307 - val_loss: 0.1566 - val_accuracy: 0.9497 - 631ms/epoch - 18ms/step
Epoch 6/20
36/36 - 1s - loss: 0.1800 - accuracy: 0.9313 - val_loss: 0.1831 - val_accuracy: 0.9466 - 604ms/epoch - 17ms/step
Epoch 7/20
36/36 - 1s - loss: 0.1712 - accuracy: 0.9322 - val_loss: 0.1245 - val_accuracy: 0.9629 - 557ms/epoch - 15ms/step
Epoch 8/20
36/36 - 1s - loss: 0.1579 - accuracy: 0.9362 - val_loss: 0.1369 - val_accuracy: 0.9548 - 563ms/epoch - 16ms/step
Epoch 9/20
36/36 - 1s - loss: 0.1509 - accuracy: 0.9364 - val_loss: 0.1813 - val_accuracy: 0.9482 - 597ms/epoch - 17ms/step
Epoch 10/20
36/36 - 1s - loss: 0.1418 - accuracy: 0.9428 - val_loss: 0.1211 - val_accuracy: 0.9533 - 932ms/epoch - 26ms/step
Epoch 11/20
36/36 - 1s - loss: 0.1501 - accuracy: 0.9381 - val_loss: 0.1408 - val_accuracy: 0.9573 - 1s/epoch - 28ms/step
Epoch 12/20
36/36 - 1s - loss: 0.1415 - accuracy: 0.9414 - val_loss: 0.1090 - val_accuracy: 0.9609 - 1s/epoch - 28ms/step
Epoch 13/20
36/36 - 1s - loss: 0.1471 - accuracy: 0.9385 - val_loss: 0.1580 - val_accuracy: 0.9502 - 991ms/epoch - 28ms/step
Epoch 14/20
36/36 - 1s - loss: 0.1431 - accuracy: 0.9392 - val_loss: 0.1121 - val_accuracy: 0.9604 - 974ms/epoch - 27ms/step
Epoch 15/20
36/36 - 1s - loss: 0.1434 - accuracy: 0.9408 - val_loss: 0.1076 - val_accuracy: 0.9604 - 1s/epoch - 30ms/step
Epoch 16/20
36/36 - 1s - loss: 0.1539 - accuracy: 0.9358 - val_loss: 0.1326 - val_accuracy: 0.9507 - 848ms/epoch - 24ms/step
Epoch 17/20
36/36 - 1s - loss: 0.1430 - accuracy: 0.9387 - val_loss: 0.1036 - val_accuracy: 0.9634 - 666ms/epoch - 19ms/step
Epoch 18/20
36/36 - 1s - loss: 0.1454 - accuracy: 0.9385 - val_loss: 0.1480 - val_accuracy: 0.9563 - 679ms/epoch - 19ms/step
Epoch 19/20
36/36 - 1s - loss: 0.1666 - accuracy: 0.9279 - val_loss: 0.1729 - val_accuracy: 0.9553 - 673ms/epoch - 19ms/step
Epoch 20/20
36/36 - 1s - loss: 0.1547 - accuracy: 0.9377 - val_loss: 0.1163 - val_accuracy: 0.9629 - 664ms/epoch - 18ms/step
Validation accuracy for Model of [Learning Rate 0.1 | Num Filters 128 | Dropout Rate 0.5 | FilterSize 5]: 0.9629064798355103
Epoch 1/20
36/36 - 2s - loss: 2.4421 - accuracy: 0.6828 - val_loss: 0.2388 - val_accuracy: 0.8908 - 2s/epoch - 50ms/step
Epoch 2/20
36/36 - 1s - loss: 0.2723 - accuracy: 0.8690 - val_loss: 0.1889 - val_accuracy: 0.9400 - 885ms/epoch - 25ms/step
Epoch 3/20
36/36 - 1s - loss: 0.2372 - accuracy: 0.8958 - val_loss: 0.1619 - val_accuracy: 0.9512 - 852ms/epoch - 24ms/step
Epoch 4/20
36/36 - 1s - loss: 0.2035 - accuracy: 0.9067 - val_loss: 0.1567 - val_accuracy: 0.9563 - 828ms/epoch - 23ms/step
Epoch 5/20
36/36 - 1s - loss: 0.1785 - accuracy: 0.9205 - val_loss: 0.1612 - val_accuracy: 0.9558 - 819ms/epoch - 23ms/step
Epoch 6/20
36/36 - 1s - loss: 0.1757 - accuracy: 0.9225 - val_loss: 0.1222 - val_accuracy: 0.9563 - 833ms/epoch - 23ms/step
Epoch 7/20
36/36 - 1s - loss: 0.1759 - accuracy: 0.9173 - val_loss: 0.1701 - val_accuracy: 0.9568 - 667ms/epoch - 19ms/step
Epoch 8/20
36/36 - 1s - loss: 0.1768 - accuracy: 0.9229 - val_loss: 0.1447 - val_accuracy: 0.9629 - 555ms/epoch - 15ms/step
Epoch 9/20
36/36 - 1s - loss: 0.1612 - accuracy: 0.9303 - val_loss: 0.1653 - val_accuracy: 0.9543 - 528ms/epoch - 15ms/step
Epoch 10/20
36/36 - 1s - loss: 0.1587 - accuracy: 0.9305 - val_loss: 0.1344 - val_accuracy: 0.9624 - 555ms/epoch - 15ms/step
Epoch 11/20
36/36 - 1s - loss: 0.1552 - accuracy: 0.9347 - val_loss: 0.1259 - val_accuracy: 0.9629 - 563ms/epoch - 16ms/step
Epoch 12/20
36/36 - 1s - loss: 0.1566 - accuracy: 0.9332 - val_loss: 0.1457 - val_accuracy: 0.9624 - 558ms/epoch - 16ms/step
Epoch 13/20
36/36 - 1s - loss: 0.1441 - accuracy: 0.9382 - val_loss: 0.1351 - val_accuracy: 0.9507 - 633ms/epoch - 18ms/step
Epoch 14/20
36/36 - 1s - loss: 0.1526 - accuracy: 0.9357 - val_loss: 0.1083 - val_accuracy: 0.9599 - 634ms/epoch - 18ms/step
Epoch 15/20
36/36 - 1s - loss: 0.1482 - accuracy: 0.9348 - val_loss: 0.1076 - val_accuracy: 0.9660 - 622ms/epoch - 17ms/step
Epoch 16/20
36/36 - 1s - loss: 0.1489 - accuracy: 0.9326 - val_loss: 0.1306 - val_accuracy: 0.9604 - 594ms/epoch - 16ms/step
Epoch 17/20
36/36 - 1s - loss: 0.1372 - accuracy: 0.9361 - val_loss: 0.1344 - val_accuracy: 0.9660 - 553ms/epoch - 15ms/step
Epoch 18/20
36/36 - 1s - loss: 0.1393 - accuracy: 0.9371 - val_loss: 0.1142 - val_accuracy: 0.9665 - 568ms/epoch - 16ms/step
Epoch 19/20
36/36 - 1s - loss: 0.1310 - accuracy: 0.9408 - val_loss: 0.1218 - val_accuracy: 0.9639 - 587ms/epoch - 16ms/step
Epoch 20/20
36/36 - 1s - loss: 0.1366 - accuracy: 0.9363 - val_loss: 0.1326 - val_accuracy: 0.9568 - 585ms/epoch - 16ms/step
Validation accuracy for Model of [Learning Rate 0.1 | Num Filters 128 | Dropout Rate 0.5 | FilterSize 7]: 0.9568089246749878
Epoch 1/20
36/36 - 2s - loss: 1.4128 - accuracy: 0.7556 - val_loss: 0.2215 - val_accuracy: 0.9273 - 2s/epoch - 50ms/step
Epoch 2/20
36/36 - 1s - loss: 0.2518 - accuracy: 0.8960 - val_loss: 0.1685 - val_accuracy: 0.9360 - 699ms/epoch - 19ms/step
Epoch 3/20
36/36 - 1s - loss: 0.2149 - accuracy: 0.9064 - val_loss: 0.1469 - val_accuracy: 0.9400 - 660ms/epoch - 18ms/step
Epoch 4/20
36/36 - 1s - loss: 0.2006 - accuracy: 0.9061 - val_loss: 0.1491 - val_accuracy: 0.9456 - 677ms/epoch - 19ms/step
Epoch 5/20
36/36 - 1s - loss: 0.1865 - accuracy: 0.9126 - val_loss: 0.1277 - val_accuracy: 0.9375 - 675ms/epoch - 19ms/step
Epoch 6/20
36/36 - 1s - loss: 0.1690 - accuracy: 0.9212 - val_loss: 0.1371 - val_accuracy: 0.9395 - 663ms/epoch - 18ms/step
Epoch 7/20
36/36 - 1s - loss: 0.1679 - accuracy: 0.9198 - val_loss: 0.1639 - val_accuracy: 0.9151 - 711ms/epoch - 20ms/step
Epoch 8/20
36/36 - 1s - loss: 0.1847 - accuracy: 0.9148 - val_loss: 0.1211 - val_accuracy: 0.9370 - 696ms/epoch - 19ms/step
Epoch 9/20
36/36 - 1s - loss: 0.1717 - accuracy: 0.9167 - val_loss: 0.1321 - val_accuracy: 0.9390 - 682ms/epoch - 19ms/step
Epoch 10/20
36/36 - 1s - loss: 0.1659 - accuracy: 0.9204 - val_loss: 0.1597 - val_accuracy: 0.9212 - 950ms/epoch - 26ms/step
Epoch 11/20
36/36 - 1s - loss: 0.1639 - accuracy: 0.9173 - val_loss: 0.1676 - val_accuracy: 0.9411 - 1s/epoch - 30ms/step
Epoch 12/20
36/36 - 1s - loss: 0.1692 - accuracy: 0.9205 - val_loss: 0.1385 - val_accuracy: 0.9553 - 1s/epoch - 32ms/step
Epoch 13/20
36/36 - 1s - loss: 0.1684 - accuracy: 0.9208 - val_loss: 0.1465 - val_accuracy: 0.9548 - 1s/epoch - 32ms/step
Epoch 14/20
36/36 - 1s - loss: 0.1647 - accuracy: 0.9198 - val_loss: 0.1368 - val_accuracy: 0.9563 - 940ms/epoch - 26ms/step
Epoch 15/20
36/36 - 1s - loss: 0.1719 - accuracy: 0.9227 - val_loss: 0.1414 - val_accuracy: 0.9482 - 673ms/epoch - 19ms/step
Epoch 16/20
36/36 - 1s - loss: 0.1668 - accuracy: 0.9187 - val_loss: 0.1444 - val_accuracy: 0.9502 - 616ms/epoch - 17ms/step
Epoch 17/20
36/36 - 1s - loss: 0.1763 - accuracy: 0.9176 - val_loss: 0.1578 - val_accuracy: 0.9466 - 603ms/epoch - 17ms/step
Epoch 18/20
36/36 - 1s - loss: 0.1663 - accuracy: 0.9205 - val_loss: 0.1289 - val_accuracy: 0.9563 - 621ms/epoch - 17ms/step
Epoch 19/20
36/36 - 1s - loss: 0.1603 - accuracy: 0.9187 - val_loss: 0.1427 - val_accuracy: 0.9578 - 572ms/epoch - 16ms/step
Epoch 20/20
36/36 - 1s - loss: 0.1639 - accuracy: 0.9156 - val_loss: 0.1684 - val_accuracy: 0.9441 - 611ms/epoch - 17ms/step
Validation accuracy for Model of [Learning Rate 0.1 | Num Filters 128 | Dropout Rate 0.7 | FilterSize 3]: 0.9441056847572327
Epoch 1/20
36/36 - 2s - loss: 1.4300 - accuracy: 0.7351 - val_loss: 0.2067 - val_accuracy: 0.9395 - 2s/epoch - 64ms/step
Epoch 2/20
36/36 - 1s - loss: 0.2693 - accuracy: 0.8771 - val_loss: 0.2020 - val_accuracy: 0.9512 - 981ms/epoch - 27ms/step
Epoch 3/20
36/36 - 1s - loss: 0.2556 - accuracy: 0.8800 - val_loss: 0.1719 - val_accuracy: 0.9563 - 1s/epoch - 29ms/step
Epoch 4/20
36/36 - 1s - loss: 0.2617 - accuracy: 0.8799 - val_loss: 0.2019 - val_accuracy: 0.9558 - 994ms/epoch - 28ms/step
Epoch 5/20
36/36 - 1s - loss: 0.2400 - accuracy: 0.8891 - val_loss: 0.1631 - val_accuracy: 0.9614 - 923ms/epoch - 26ms/step
Epoch 6/20
36/36 - 1s - loss: 0.2676 - accuracy: 0.8739 - val_loss: 0.1797 - val_accuracy: 0.9553 - 712ms/epoch - 20ms/step
Epoch 7/20
36/36 - 1s - loss: 0.2636 - accuracy: 0.8758 - val_loss: 0.1866 - val_accuracy: 0.9472 - 626ms/epoch - 17ms/step
Epoch 8/20
36/36 - 1s - loss: 0.2589 - accuracy: 0.8754 - val_loss: 0.1677 - val_accuracy: 0.9553 - 621ms/epoch - 17ms/step
Epoch 9/20
36/36 - 1s - loss: 0.2587 - accuracy: 0.8786 - val_loss: 0.1830 - val_accuracy: 0.9477 - 622ms/epoch - 17ms/step
Epoch 10/20
36/36 - 1s - loss: 0.2415 - accuracy: 0.8851 - val_loss: 0.1807 - val_accuracy: 0.9553 - 654ms/epoch - 18ms/step
Epoch 11/20
36/36 - 1s - loss: 0.2476 - accuracy: 0.8801 - val_loss: 0.1692 - val_accuracy: 0.9568 - 653ms/epoch - 18ms/step
Epoch 12/20
36/36 - 1s - loss: 0.2944 - accuracy: 0.8615 - val_loss: 0.2192 - val_accuracy: 0.9497 - 645ms/epoch - 18ms/step
Epoch 13/20
36/36 - 1s - loss: 0.2766 - accuracy: 0.8481 - val_loss: 0.1846 - val_accuracy: 0.9563 - 626ms/epoch - 17ms/step
Epoch 14/20
36/36 - 1s - loss: 0.2544 - accuracy: 0.8674 - val_loss: 0.1575 - val_accuracy: 0.9578 - 664ms/epoch - 18ms/step
Epoch 15/20
36/36 - 1s - loss: 0.2423 - accuracy: 0.8796 - val_loss: 0.1345 - val_accuracy: 0.9578 - 660ms/epoch - 18ms/step
Epoch 16/20
36/36 - 1s - loss: 0.2308 - accuracy: 0.8879 - val_loss: 0.1502 - val_accuracy: 0.9487 - 600ms/epoch - 17ms/step
Epoch 17/20
36/36 - 1s - loss: 0.2285 - accuracy: 0.8868 - val_loss: 0.1728 - val_accuracy: 0.9578 - 592ms/epoch - 16ms/step
Epoch 18/20
36/36 - 1s - loss: 0.2775 - accuracy: 0.8756 - val_loss: 0.1718 - val_accuracy: 0.9558 - 567ms/epoch - 16ms/step
Epoch 19/20
36/36 - 1s - loss: 0.2707 - accuracy: 0.8669 - val_loss: 0.2028 - val_accuracy: 0.9507 - 751ms/epoch - 21ms/step
Epoch 20/20
36/36 - 1s - loss: 0.3048 - accuracy: 0.8285 - val_loss: 0.2362 - val_accuracy: 0.9522 - 719ms/epoch - 20ms/step
Validation accuracy for Model of [Learning Rate 0.1 | Num Filters 128 | Dropout Rate 0.7 | FilterSize 5]: 0.952235758304596
Epoch 1/20
36/36 - 2s - loss: 2.2991 - accuracy: 0.6856 - val_loss: 0.2051 - val_accuracy: 0.9284 - 2s/epoch - 46ms/step
Epoch 2/20
36/36 - 1s - loss: 0.3057 - accuracy: 0.8430 - val_loss: 0.1972 - val_accuracy: 0.8918 - 584ms/epoch - 16ms/step
Epoch 3/20
36/36 - 1s - loss: 0.2360 - accuracy: 0.8842 - val_loss: 0.1901 - val_accuracy: 0.9197 - 539ms/epoch - 15ms/step
Epoch 4/20
36/36 - 0s - loss: 0.2210 - accuracy: 0.8900 - val_loss: 0.1281 - val_accuracy: 0.9405 - 495ms/epoch - 14ms/step
Epoch 5/20
36/36 - 0s - loss: 0.1977 - accuracy: 0.9032 - val_loss: 0.1301 - val_accuracy: 0.9446 - 489ms/epoch - 14ms/step
Epoch 6/20
36/36 - 1s - loss: 0.1853 - accuracy: 0.9188 - val_loss: 0.1309 - val_accuracy: 0.9390 - 531ms/epoch - 15ms/step
Epoch 7/20
36/36 - 0s - loss: 0.1779 - accuracy: 0.9125 - val_loss: 0.1568 - val_accuracy: 0.9319 - 490ms/epoch - 14ms/step
Epoch 8/20
36/36 - 1s - loss: 0.1844 - accuracy: 0.9167 - val_loss: 0.1254 - val_accuracy: 0.9456 - 503ms/epoch - 14ms/step
Epoch 9/20
36/36 - 1s - loss: 0.1830 - accuracy: 0.9193 - val_loss: 0.1544 - val_accuracy: 0.9309 - 526ms/epoch - 15ms/step
Epoch 10/20
36/36 - 1s - loss: 0.1835 - accuracy: 0.9202 - val_loss: 0.1677 - val_accuracy: 0.9238 - 517ms/epoch - 14ms/step
Epoch 11/20
36/36 - 1s - loss: 0.1793 - accuracy: 0.9009 - val_loss: 0.1373 - val_accuracy: 0.9324 - 535ms/epoch - 15ms/step
Epoch 12/20
36/36 - 1s - loss: 0.1679 - accuracy: 0.9090 - val_loss: 0.1702 - val_accuracy: 0.9365 - 509ms/epoch - 14ms/step
Epoch 13/20
36/36 - 1s - loss: 0.1734 - accuracy: 0.9168 - val_loss: 0.1304 - val_accuracy: 0.9355 - 510ms/epoch - 14ms/step
Epoch 14/20
36/36 - 1s - loss: 0.1660 - accuracy: 0.9215 - val_loss: 0.1574 - val_accuracy: 0.9365 - 705ms/epoch - 20ms/step
Epoch 15/20
36/36 - 1s - loss: 0.1906 - accuracy: 0.9194 - val_loss: 0.1706 - val_accuracy: 0.9253 - 811ms/epoch - 23ms/step
Epoch 16/20
36/36 - 1s - loss: 0.1937 - accuracy: 0.9035 - val_loss: 0.1697 - val_accuracy: 0.9273 - 803ms/epoch - 22ms/step
Epoch 17/20
36/36 - 1s - loss: 0.1791 - accuracy: 0.9114 - val_loss: 0.1353 - val_accuracy: 0.9345 - 808ms/epoch - 22ms/step
Epoch 18/20
36/36 - 1s - loss: 0.1833 - accuracy: 0.9188 - val_loss: 0.1636 - val_accuracy: 0.9172 - 808ms/epoch - 22ms/step
Epoch 19/20
36/36 - 1s - loss: 0.1759 - accuracy: 0.9127 - val_loss: 0.1476 - val_accuracy: 0.9375 - 800ms/epoch - 22ms/step
Epoch 20/20
36/36 - 1s - loss: 0.1963 - accuracy: 0.9093 - val_loss: 0.1470 - val_accuracy: 0.9334 - 597ms/epoch - 17ms/step
Validation accuracy for Model of [Learning Rate 0.1 | Num Filters 128 | Dropout Rate 0.7 | FilterSize 7]: 0.9334349632263184

print("Best Model Validation Accuracy: " + str(best_val_accuracy))
print("Best Model Parameters: ")
print(best_model_parameters)

Best Model Validation Accuracy: 0.9766260385513306
Best Model Parameters: 
{'learning_rate': 0.01, 'num_filters': 128, 'droupout_rate': 0.3, 'filter_size': 7}

# Train the best model on the full training data
best_model.fit(X_train_res, y_train_res, epochs=num_epochs, batch_size=batch_size, verbose=1)

# Evaluate the best model on the test set
y_pred = best_model.predict(X_test)
y_pred_binary = (y_pred > 0.5).astype(int)  # Convert probabilities to binary predictions (0 or 1)

accuracy = accuracy_score(y_test, y_pred_binary)
precision = precision_score(y_test, y_pred_binary)
recall = recall_score(y_test, y_pred_binary)
f1 = f1_score(y_test, y_pred_binary)
roc_auc = roc_auc_score(y_test, y_pred_binary)
conf_matrix = confusion_matrix(y_test, y_pred_binary)

print(f"Accuracy: {accuracy}")
print(f"Precision: {precision}")
print(f"Recall: {recall}")
print(f"F1 Score: {f1}")
print(f"ROC AUC Score: {roc_auc}")
print(f"Confusion Matrix:\n{conf_matrix}")

Epoch 1/20
36/36 [==============================] - 1s 14ms/step - loss: 0.0505 - accuracy: 0.9811
Epoch 2/20
36/36 [==============================] - 1s 15ms/step - loss: 0.0561 - accuracy: 0.9788
Epoch 3/20
36/36 [==============================] - 0s 14ms/step - loss: 0.0485 - accuracy: 0.9814
Epoch 4/20
36/36 [==============================] - 1s 15ms/step - loss: 0.0460 - accuracy: 0.9818
Epoch 5/20
36/36 [==============================] - 1s 14ms/step - loss: 0.0433 - accuracy: 0.9823
Epoch 6/20
36/36 [==============================] - 1s 15ms/step - loss: 0.0458 - accuracy: 0.9826
Epoch 7/20
36/36 [==============================] - 1s 14ms/step - loss: 0.0443 - accuracy: 0.9822
Epoch 8/20
36/36 [==============================] - 1s 17ms/step - loss: 0.0434 - accuracy: 0.9822
Epoch 9/20
36/36 [==============================] - 1s 16ms/step - loss: 0.0395 - accuracy: 0.9848
Epoch 10/20
36/36 [==============================] - 1s 16ms/step - loss: 0.0419 - accuracy: 0.9832
Epoch 11/20
36/36 [==============================] - 1s 16ms/step - loss: 0.0419 - accuracy: 0.9828
Epoch 12/20
36/36 [==============================] - 1s 15ms/step - loss: 0.0377 - accuracy: 0.9846
Epoch 13/20
36/36 [==============================] - 1s 16ms/step - loss: 0.0382 - accuracy: 0.9855
Epoch 14/20
36/36 [==============================] - 1s 16ms/step - loss: 0.0422 - accuracy: 0.9829
Epoch 15/20
36/36 [==============================] - 1s 16ms/step - loss: 0.0450 - accuracy: 0.9825
Epoch 16/20
36/36 [==============================] - 1s 16ms/step - loss: 0.0421 - accuracy: 0.9829
Epoch 17/20
36/36 [==============================] - 1s 17ms/step - loss: 0.0407 - accuracy: 0.9848
Epoch 18/20
36/36 [==============================] - 1s 26ms/step - loss: 0.0445 - accuracy: 0.9812
Epoch 19/20
36/36 [==============================] - 1s 25ms/step - loss: 0.0432 - accuracy: 0.9827
Epoch 20/20
36/36 [==============================] - 1s 24ms/step - loss: 0.0391 - accuracy: 0.9833
31/31 [==============================] - 0s 5ms/step
Accuracy: 0.9736040609137055
Precision: 0.9571428571428572
Recall: 0.9220183486238532
F1 Score: 0.9392523364485982
ROC AUC Score: 0.9551421599703361
Confusion Matrix:
[[758   9]
 [ 17 201]]

# Visualising the accuracy and loss graph for the model
history = best_model.fit(X_train_res, y_train_res, epochs=20, batch_size=300, validation_data=(X_val, y_val), verbose=0)
plt.plot(history.history['accuracy'])
plt.plot(history.history['val_accuracy'])
plt.title('model accuracy')
plt.ylabel('accuracy')
plt.xlabel('epoch')
plt.legend(['train', 'val'], loc='upper left')
plt.show()

plt.plot(history.history['loss'])
plt.plot(history.history['val_loss'])
plt.title('model loss')
plt.ylabel('loss')
plt.xlabel('epoch')
plt.legend(['train', 'val'], loc='upper left')
plt.show()

# Run SHAP explainer on a subset of training data because running on entire training set will take very long
explainer = shap.DeepExplainer(best_model,shap.sample(X_train_res, 100))

# Check the shape of the current training set
X_train_res.shape

(10726, 21, 1)

# Check the shape of the current test set
X_test.shape

(985, 21, 1)

# Get shap values for each point in dataset
shap_values = explainer.shap_values(X_test)

# Check that the shape of the shap_values list is the same as that of the test set
shap_values[0].shape

(985, 21, 1)

# Both training and test set are three-dim because that is how the model receives its input data as a 1D sequence
# However, to visualize the impact of variables on a plot, we require the data to be in 2D
# We know there are 985 obs and 21 features in th test set, meaning we should flatten the third dimension
X_test_flatten = X_test.reshape(985, 21)

# Check the shape of test set
X_test_flatten.shape

(985, 21)

# The same explanation is applied to the shap values
shap_values_flatten = shap_values[0].reshape(985, 21)
# Check the shape of shap values
shap_values_flatten.shape

(985, 21)

shap.initjs()
ind = 16
print(y_test.iloc[ind])
shap.force_plot(
    explainer.expected_value.numpy(), shap_values_flatten[ind,:], X_test_flatten[ind,:],
    feature_names=X_train.columns
)

0

shap.initjs()
ind = 933
print(y_test.iloc[ind])
shap.force_plot(
    explainer.expected_value.numpy(), shap_values_flatten[ind,:], X_test_flatten[ind,:],
    feature_names=X_train.columns
)

1

shap.initjs()
shap.summary_plot(shap_values_flatten, X_test_flatten, feature_names=X_train.columns)

	mean	std	min	median	max	variance	mode	mad	kurtosis	skewness
Y	2.214206e-01	4.152241e-01	0.00	0.000000	1.000000e+00	1.724110e-01	0.0	0.000000	-0.199318	1.341895
Avg min between sent tnx	5.086879e+03	2.148655e+04	0.00	17.340000	4.302877e+05	4.616718e+08	0.0	17.340000	95.032225	8.418716
Avg min between received tnx	8.004851e+03	2.308171e+04	0.00	509.770000	4.821755e+05	5.327656e+08	0.0	509.770000	67.971030	6.744270
Time Diff between first and last (Mins)	2.183333e+05	3.229379e+05	0.00	46637.030000	1.954861e+06	1.042889e+11	0.0	46637.030000	2.953471	1.809701
Sent tnx	1.159317e+02	7.572264e+02	0.00	3.000000	1.000000e+04	5.733918e+05	0.0	3.000000	120.634073	10.482946
Received Tnx	1.637009e+02	9.408366e+02	0.00	4.000000	1.000000e+04	8.851734e+05	1.0	3.000000	82.576773	8.820039
Number of Created Contracts	3.729702e+00	1.414456e+02	0.00	0.000000	9.995000e+03	2.000685e+04	0.0	0.000000	3112.610930	51.712336
Unique Received From Addresses	3.036094e+01	2.986211e+02	0.00	2.000000	9.999000e+03	8.917457e+04	1.0	1.000000	418.567105	18.113345
Unique Sent To Addresses	2.584016e+01	2.638204e+02	0.00	2.000000	9.287000e+03	6.960121e+04	1.0	1.000000	410.105291	18.351527
min value received	4.384515e+01	3.259291e+02	0.00	0.095856	1.000000e+04	1.062298e+05	0.0	0.095856	657.800891	23.292332
max value received	5.231525e+02	1.300882e+04	0.00	6.000000	8.000000e+05	1.692294e+08	101.0	5.978713	2386.259591	46.416606
avg val received	1.007117e+02	2.885002e+03	0.00	1.729730	2.836188e+05	8.323238e+06	101.0	1.713606	9476.688098	96.498968
min val sent	4.800090e+00	1.386097e+02	0.00	0.049126	1.200000e+04	1.921264e+04	0.0	0.049126	5950.739030	73.415556
max val sent	3.146173e+02	6.629213e+03	0.00	4.999380	5.200000e+05	4.394646e+07	0.0	4.999380	4190.770883	59.824593
avg val sent	4.475573e+01	2.390802e+02	0.00	1.606000	1.200000e+04	5.715935e+04	0.0	1.606000	1036.761326	25.527195
total transactions (including tnx to create contract	2.833624e+02	1.352404e+03	0.00	8.000000	1.999500e+04	1.828997e+06	4.0	6.000000	50.006440	6.848002
total Ether sent	1.016092e+04	3.583227e+05	0.00	12.486800	2.858096e+07	1.283952e+11	0.0	12.486800	4511.035091	62.353731
total ether received	1.163832e+04	3.642048e+05	0.00	30.529634	2.858159e+07	1.326451e+11	101.0	30.518634	4151.038376	58.786322
total ether balance	1.477395e+03	2.424254e+05	-15605352.04	0.001722	1.428864e+07	5.877009e+10	0.0	0.002324	3134.128893	-1.205078
Total ERC20 tnxs	3.320150e+01	4.283809e+02	0.00	0.000000	1.000100e+04	1.835102e+05	0.0	0.000000	457.591042	20.827358
ERC20 total Ether received	1.187015e+08	1.008496e+10	0.00	0.000000	1.000020e+12	1.017063e+20	0.0	0.000000	9818.588541	99.056107
ERC20 total ether sent	1.270022e+07	1.129580e+09	0.00	0.000000	1.120000e+11	1.275951e+18	0.0	0.000000	9815.753901	99.035567
ERC20 total Ether sent contract	1.015938e+02	5.864874e+03	0.00	0.000000	4.160000e+05	3.439675e+07	0.0	0.000000	3969.297666	61.938745
ERC20 uniq sent addr	5.163093e+00	1.007335e+02	0.00	0.000000	6.582000e+03	1.014723e+04	0.0	0.000000	2287.864646	42.468140
ERC20 uniq rec addr	6.958439e+00	7.832457e+01	0.00	0.000000	4.293000e+03	6.134739e+03	0.0	0.000000	1913.320348	39.251875
ERC20 uniq rec contract addr	4.488975e+00	1.656017e+01	0.00	0.000000	7.820000e+02	2.742391e+02	0.0	0.000000	592.948783	16.962601
ERC20 min val rec	4.447068e+02	1.615701e+04	0.00	0.000000	9.900000e+05	2.610488e+08	0.0	0.000000	3040.139062	52.802317
ERC20 max val rec	1.147012e+08	1.008383e+10	0.00	0.000000	1.000000e+12	1.016835e+20	0.0	0.000000	9822.363770	99.084581
ERC20 avg val rec	3.980082e+06	2.049048e+08	0.00	0.000000	1.724181e+10	4.198599e+16	0.0	0.000000	5787.507651	74.211009
ERC20 min val sent	1.075218e+04	1.008216e+06	0.00	0.000000	1.000000e+08	1.016499e+12	0.0	0.000000	9829.151022	99.135173
ERC20 max val sent	1.193780e+07	1.129115e+09	0.00	0.000000	1.120000e+11	1.274901e+18	0.0	0.000000	9832.192507	99.158044
ERC20 avg val sent	5.786132e+06	5.660157e+08	0.00	0.000000	5.614756e+10	3.203738e+17	0.0	0.000000	9834.386402	99.174560
ERC20 uniq sent token name	1.268265e+00	6.456639e+00	0.00	0.000000	2.130000e+02	4.168819e+01	0.0	0.000000	275.797972	12.841933
ERC20 uniq rec token name	4.420079e+00	1.601679e+01	0.00	0.000000	7.370000e+02	2.565375e+02	0.0	0.000000	541.546339	16.138336
ERC20_most_sent_token_valid_name	5.528910e-01	4.972199e-01	0.00	1.000000	1.000000e+00	2.472277e-01	1.0	0.000000	-1.954734	-0.212758
ERC20_most_rec_token_valid_name	5.529926e-01	4.972091e-01	0.00	1.000000	1.000000e+00	2.472169e-01	1.0	0.000000	-1.954558	-0.213171

BT4012 Team 19 Group Project¶

Team members¶

1. Importing the required libraries¶

2. Loading Dataset¶

3. Data Cleaning¶

Irrelevant and Errorneous Columns¶

Missing values¶

Low Variability¶

Outliers¶

4. EDA¶

Fraud Ratio¶

Feature Correlation and Treament¶

Distribution of Features Partitioned by Target Variable¶

5. Feature Engineering¶

Feature Creation¶

6. Data Pre-processing¶

Global Preprocessing: Handling Skewed Data¶

Splitting of Dataset¶

Local Preprocessing: Normalisation¶

7. Data Augmentation¶

8. Model Architecture¶

Classical Models¶

SVM¶

SVM Model¶

SVM Hyperparameter Tuning with GridSearchCV¶

Random Forest¶

Random Forest Model¶

Random Forest Hyperparameter Tuning with RandomSearchCV¶

Random Forest Hyperparameter Tuning with GridSearchCV¶

Model: CNN¶

9. Model Tuning¶

10. Results and Discussion¶

Model Results¶

Model Interpretability: Explainable AI (XAI)¶

	Unnamed: 0	Index	Address	Avg min between sent tnx	Avg min between received tnx	Time Diff between first and last (Mins)	Sent tnx	Received Tnx	Number of Created Contracts	...	ERC20 min val sent	ERC20 max val sent	ERC20 avg val sent	ERC20 uniq sent token name	ERC20 uniq rec token name	ERC20 most sent token type	ERC20_most_rec_token_type
0	0	1	0x00009277775ac7d0d59eaad8fee3d10ac6c805e8	844.26	1093.71	704785.63	721	89	0	...	0.000000	1.683100e+07	271779.920000	39.0	57.0	Cofoundit	Numeraire
1	1	2	0x0002b44ddb1476db43c868bd494422ee4c136fed	12709.07	2958.44	1218216.73	94	8	0	...	2.260809	2.260809e+00	2.260809	1.0	7.0	Livepeer Token	Livepeer Token
2	2	3	0x0002bda54cb772d040f779e88eb453cac0daa244	246194.54	2434.02	516729.30	2	10	0	...	0.000000	0.000000e+00	0.000000	0.0	8.0	None	XENON
3	3	4	0x00038e6ba2fd5c09aedb96697c8d7b8fa6632e5e	10219.60	15785.09	397555.90	25	9	0	...	100.000000	9.029231e+03	3804.076893	1.0	11.0	Raiden	XENON
4	4	5	0x00062d1dd1afb6fb02540ddad9cdebfe568e0d89	36.61	10707.77	382472.42	4598	20	1	...	0.000000	4.500000e+04	13726.659220	6.0	27.0	StatusNetwork	EOS

	ERC20 max val rec	ERC20 uniq sent addr	ERC20 uniq rec addr	ERC20 uniq sent contract addr	ERC20 uniq rec contract addr	ERC20 avg time between sent tnx	ERC20 avg time between rec tnx	ERC20 avg time between contract tnx	ERC20 min val rec	ERC20 avg val rec	...	ERC20 max val sent	ERC20 avg val sent	ERC20 min val sent contract	ERC20 max val sent contract	ERC20 avg val sent contract	ERC20 uniq sent token name	ERC20 uniq rec token name	ERC20 total Ether sent contract	ERC20 total Ether received	Total ERC20 tnxs
7662	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
7666	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
7675	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
7676	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
7678	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
9831	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
9833	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
9834	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
9835	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
9839	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN

	Y	Avg min between sent tnx	Avg min between received tnx	Time Diff between first and last (Mins)	Sent tnx	Received Tnx	Number of Created Contracts	Unique Received From Addresses	Unique Sent To Addresses	min value received	max value received	avg val received	min val sent	max val sent	avg val sent	total transactions (including tnx to create contract	total Ether sent	total ether received	total ether balance	Total ERC20 tnxs	ERC20 total Ether received	ERC20 total ether sent	ERC20 uniq sent addr	ERC20 uniq rec addr	ERC20 uniq rec contract addr	ERC20 min val rec	ERC20 max val rec	ERC20 avg val rec	ERC20 min val sent	ERC20 max val sent	ERC20 avg val sent	ERC20 uniq sent token name	ERC20 uniq rec token name
Y	1.00	-0.03	-0.12	-0.27	-0.08	-0.08	-0.01	-0.03	-0.05	-0.02	-0.02	-0.01	0.01	-0.02	-0.06	-0.10	-0.01	-0.02	-0.00	-0.03	-0.01	0.02	-0.03	-0.03	-0.06	0.00	-0.01	0.00	0.02	0.02	0.02	-0.03	-0.06
Avg min between sent tnx	-0.03	1.00	0.06	0.21	-0.03	-0.04	-0.01	-0.02	-0.02	-0.01	-0.01	-0.00	-0.00	-0.01	0.00	-0.04	-0.01	-0.01	-0.00	-0.01	-0.00	-0.00	-0.01	0.00	0.05	0.00	-0.00	-0.00	-0.00	-0.00	-0.00	0.00	0.05
Avg min between received tnx	-0.12	0.06	1.00	0.30	-0.04	-0.05	-0.01	-0.03	-0.03	-0.05	-0.01	-0.01	-0.01	-0.01	-0.04	-0.06	-0.01	-0.01	-0.00	-0.02	-0.00	-0.00	-0.01	-0.01	-0.01	-0.01	-0.00	-0.01	-0.00	-0.00	-0.00	-0.02	-0.01
Time Diff between first and last (Mins)	-0.27	0.21	0.30	1.00	0.15	0.15	-0.00	0.04	0.07	-0.08	-0.00	-0.01	-0.01	0.01	-0.05	0.19	0.01	0.01	0.00	0.08	0.05	-0.00	0.04	0.08	0.33	-0.01	0.05	0.05	-0.01	-0.01	-0.01	0.27	0.33
Sent tnx	-0.08	-0.03	-0.04	0.15	1.00	0.20	0.32	0.13	0.67	0.02	0.10	0.14	-0.00	0.23	0.03	0.73	0.24	0.16	-0.13	0.38	0.01	-0.00	0.36	0.30	0.22	-0.00	0.00	0.01	-0.00	-0.00	-0.00	0.08	0.22
Received Tnx	-0.08	-0.04	-0.05	0.15	0.20	1.00	-0.00	0.65	0.16	-0.02	0.22	-0.00	0.09	0.10	0.13	0.81	0.13	0.24	0.16	0.12	0.02	-0.00	0.04	0.14	0.20	-0.00	0.02	0.02	-0.00	-0.00	-0.00	0.05	0.21
Number of Created Contracts	-0.01	-0.01	-0.01	-0.00	0.32	-0.00	1.00	-0.00	0.08	-0.00	-0.00	-0.00	-0.00	0.14	-0.00	0.28	0.02	-0.00	-0.04	0.25	0.00	0.00	0.15	0.19	0.03	-0.00	0.00	0.00	-0.00	0.00	-0.00	0.01	0.03
Unique Received From Addresses	-0.03	-0.02	-0.03	0.04	0.13	0.65	-0.00	1.00	0.16	-0.01	0.18	-0.00	0.30	0.06	0.23	0.52	0.03	0.12	0.14	0.06	0.00	0.00	0.05	0.08	0.15	-0.00	0.00	0.00	-0.00	-0.00	-0.00	0.04	0.15
Unique Sent To Addresses	-0.05	-0.02	-0.03	0.07	0.67	0.16	0.08	0.16	1.00	0.07	0.15	0.21	-0.00	0.20	0.02	0.50	0.16	0.09	-0.11	0.15	0.01	0.00	0.12	0.18	0.24	-0.00	0.00	0.01	-0.00	-0.00	-0.00	0.09	0.24
min value received	-0.02	-0.01	-0.05	-0.08	0.02	-0.02	-0.00	-0.01	0.07	1.00	0.03	0.12	0.12	0.02	0.27	-0.00	-0.00	-0.00	-0.00	-0.01	-0.00	-0.00	-0.01	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.03	0.00
max value received	-0.02	-0.01	-0.01	-0.00	0.10	0.22	-0.00	0.18	0.15	0.03	1.00	0.62	0.00	0.14	0.04	0.21	0.11	0.30	0.28	0.03	0.01	-0.00	0.00	0.04	0.18	-0.00	0.00	0.00	-0.00	-0.00	-0.00	0.01	0.18
avg val received	-0.01	-0.00	-0.01	-0.01	0.14	-0.00	-0.00	-0.00	0.21	0.12	0.62	1.00	0.01	0.13	0.07	0.08	0.16	0.06	-0.14	0.01	0.00	-0.00	0.00	0.04	0.20	-0.00	0.00	-0.00	-0.00	-0.00	-0.00	0.01	0.20
min val sent	0.01	-0.00	-0.01	-0.01	-0.00	0.09	-0.00	0.30	-0.00	0.12	0.00	0.01	1.00	0.02	0.59	0.06	0.00	0.00	0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.01	-0.00
max val sent	-0.02	-0.01	-0.01	0.01	0.23	0.10	0.14	0.06	0.20	0.02	0.14	0.13	0.02	1.00	0.18	0.21	0.47	0.09	-0.56	0.13	0.03	0.00	0.05	0.20	0.17	-0.00	0.03	0.03	-0.00	-0.00	-0.00	0.02	0.17
avg val sent	-0.06	0.00	-0.04	-0.05	0.03	0.13	-0.00	0.23	0.02	0.27	0.04	0.07	0.59	0.18	1.00	0.10	0.20	0.17	-0.05	0.02	0.02	-0.00	-0.01	0.01	0.05	-0.00	0.02	0.01	-0.00	-0.00	-0.00	-0.02	0.05
total transactions (including tnx to create contract	-0.10	-0.04	-0.06	0.19	0.73	0.81	0.28	0.52	0.50	-0.00	0.21	0.08	0.06	0.21	0.10	1.00	0.23	0.25	0.03	0.32	0.02	-0.00	0.25	0.28	0.27	-0.00	0.01	0.02	-0.00	-0.00	-0.00	0.08	0.27
total Ether sent	-0.01	-0.01	-0.01	0.01	0.24	0.13	0.02	0.03	0.16	-0.00	0.11	0.16	0.00	0.47	0.20	0.23	1.00	0.77	-0.31	0.07	0.00	0.00	0.01	0.04	0.09	-0.00	0.00	0.00	-0.00	-0.00	-0.00	0.01	0.09
total ether received	-0.02	-0.01	-0.01	0.01	0.16	0.24	-0.00	0.12	0.09	-0.00	0.30	0.06	0.00	0.09	0.17	0.25	0.77	1.00	0.36	0.07	0.00	-0.00	0.00	0.03	0.07	-0.00	0.00	0.00	-0.00	-0.00	-0.00	0.00	0.07
total ether balance	-0.00	-0.00	-0.00	0.00	-0.13	0.16	-0.04	0.14	-0.11	-0.00	0.28	-0.14	0.00	-0.56	-0.05	0.03	-0.31	0.36	1.00	-0.01	-0.00	-0.00	-0.01	-0.02	-0.02	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.01	-0.02
Total ERC20 tnxs	-0.03	-0.01	-0.02	0.08	0.38	0.12	0.25	0.06	0.15	-0.01	0.03	0.01	-0.00	0.13	0.02	0.32	0.07	0.07	-0.01	1.00	0.00	0.00	0.73	0.72	0.27	-0.00	0.00	0.00	-0.00	-0.00	-0.00	0.19	0.26
ERC20 total Ether received	-0.01	-0.00	-0.00	0.05	0.01	0.02	0.00	0.00	0.01	-0.00	0.01	0.00	-0.00	0.03	0.02	0.02	0.00	0.00	-0.00	0.00	1.00	0.00	0.00	0.01	0.03	-0.00	1.00	0.86	-0.00	-0.00	-0.00	0.02	0.03
ERC20 total ether sent	0.02	-0.00	-0.00	-0.00	-0.00	-0.00	0.00	0.00	0.00	-0.00	-0.00	-0.00	-0.00	0.00	-0.00	-0.00	0.00	-0.00	-0.00	0.00	0.00	1.00	0.00	0.00	0.00	-0.00	0.00	0.00	1.00	1.00	1.00	0.00	0.00
ERC20 uniq sent addr	-0.03	-0.01	-0.01	0.04	0.36	0.04	0.15	0.05	0.12	-0.01	0.00	0.00	-0.00	0.05	-0.01	0.25	0.01	0.00	-0.01	0.73	0.00	0.00	1.00	0.57	0.15	-0.00	0.00	0.00	-0.00	-0.00	-0.00	0.11	0.14
ERC20 uniq rec addr	-0.03	0.00	-0.01	0.08	0.30	0.14	0.19	0.08	0.18	-0.00	0.04	0.04	-0.00	0.20	0.01	0.28	0.04	0.03	-0.02	0.72	0.01	0.00	0.57	1.00	0.44	-0.00	0.01	0.00	-0.00	-0.00	-0.00	0.32	0.44
ERC20 uniq rec contract addr	-0.06	0.05	-0.01	0.33	0.22	0.20	0.03	0.15	0.24	-0.00	0.18	0.20	-0.00	0.17	0.05	0.27	0.09	0.07	-0.02	0.27	0.03	0.00	0.15	0.44	1.00	-0.01	0.03	0.02	-0.00	-0.00	-0.00	0.79	1.00
ERC20 min val rec	0.00	0.00	-0.01	-0.01	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.01	1.00	-0.00	-0.00	0.01	-0.00	-0.00	-0.00	-0.01
ERC20 max val rec	-0.01	-0.00	-0.00	0.05	0.00	0.02	0.00	0.00	0.00	-0.00	0.00	0.00	-0.00	0.03	0.02	0.01	0.00	0.00	-0.00	0.00	1.00	0.00	0.00	0.01	0.03	-0.00	1.00	0.86	-0.00	-0.00	-0.00	0.02	0.03
ERC20 avg val rec	0.00	-0.00	-0.01	0.05	0.01	0.02	0.00	0.00	0.01	-0.00	0.00	-0.00	-0.00	0.03	0.01	0.02	0.00	0.00	-0.00	0.00	0.86	0.00	0.00	0.00	0.02	-0.00	0.86	1.00	-0.00	-0.00	-0.00	0.01	0.02
ERC20 min val sent	0.02	-0.00	-0.00	-0.01	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	1.00	-0.00	-0.00	-0.00	0.01	-0.00	-0.00	1.00	1.00	1.00	-0.00	-0.00
ERC20 max val sent	0.02	-0.00	-0.00	-0.01	-0.00	-0.00	0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	1.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	1.00	1.00	1.00	0.00	-0.00
ERC20 avg val sent	0.02	-0.00	-0.00	-0.01	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	1.00	-0.00	-0.00	-0.00	-0.00	-0.00	-0.00	1.00	1.00	1.00	-0.00	-0.00
ERC20 uniq sent token name	-0.03	0.00	-0.02	0.27	0.08	0.05	0.01	0.04	0.09	-0.03	0.01	0.01	-0.01	0.02	-0.02	0.08	0.01	0.00	-0.01	0.19	0.02	0.00	0.11	0.32	0.79	-0.00	0.02	0.01	-0.00	0.00	-0.00	1.00	0.79
ERC20 uniq rec token name	-0.06	0.05	-0.01	0.33	0.22	0.21	0.03	0.15	0.24	0.00	0.18	0.20	-0.00	0.17	0.05	0.27	0.09	0.07	-0.02	0.26	0.03	0.00	0.14	0.44	1.00	-0.01	0.03	0.02	-0.00	-0.00	-0.00	0.79	1.00

	Y	Avg min between sent tnx	Avg min between received tnx	Time Diff between first and last (Mins)	Number of Created Contracts	min value received	max value received	min val sent	max val sent	avg val sent	...	ERC20 total ether sent	ERC20 uniq sent addr	ERC20 uniq rec addr	ERC20 min val rec	ERC20 uniq sent token name	ERC20 uniq rec token name	ERC20_most_sent_token_valid_name	ERC20_most_rec_token_valid_name	Sent_to_Received_Ratio	Ether_Sent_to_Received_Ratio
0	0	6.739644	6.998245	13.465650	0.000000	0.000000	3.846028	0.000000	3.472587	0.788767	...	17.387945	3.433987	4.007333	0.000000	3.688879	4.060443	1	1	2.208398	0.906690
1	0	9.450150	7.992755	14.012899	0.000000	0.000000	1.284613	0.000000	1.029619	0.032316	...	1.181975	0.693147	1.791759	0.000000	0.693147	2.079442	1	1	2.545531	0.693442
2	0	12.413881	7.797710	13.155276	0.000000	0.107166	0.772630	0.048790	1.512622	1.027584	...	0.000000	0.000000	2.079442	0.000000	0.000000	2.197225	1	1	0.182322	0.693086
3	0	9.232161	9.666884	12.893093	0.000000	0.000000	6.216606	0.000000	6.111467	4.262706	...	9.342529	1.098612	2.484907	0.000000	0.693147	2.484907	1	1	1.329136	1.083325
4	0	3.627270	9.278818	12.854414	0.693147	0.000000	2.624843	0.000000	2.302585	0.022434	...	11.724328	1.609438	3.178054	0.000000	1.945910	3.332205	1	1	5.441985	1.082732
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
9836	1	9.444313	6.449506	10.981038	0.000000	0.004074	2.564949	0.382170	2.580217	2.319085	...	0.000000	0.000000	1.098612	0.000000	0.000000	1.098612	1	1	0.268264	0.693034
9837	1	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	...	0.000000	0.000000	0.693147	2.665143	0.000000	0.693147	1	1	0.693147	0.693147
9838	1	7.824222	7.691789	12.474583	0.000000	0.001077	3.105035	0.003992	2.351375	0.653459	...	0.000000	0.000000	1.791759	0.000000	0.000000	1.791759	1	1	0.939280	0.772065
9839	1	0.000000	0.000000	0.000000	0.000000	0.405465	0.405465	0.000000	0.000000	0.000000	...	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	1	1	0.000000	0.000000
9840	1	10.525238	5.014362	13.416254	0.000000	0.585135	9.852142	0.641854	6.908755	6.469913	...	0.000000	0.000000	3.637586	0.000000	0.000000	3.761200	1	1	1.945910	0.476555